GEORGIA DOT RESEARCH PROJECT 10-24 FINAL REPORT LIFE CYCLE COST ASSESSMENT AND PERFORMANCE EVALUATION OF SEDIMENT CONTROL TECHNOLOGIES OFFICE OF RESEARCH 15 Kennedy Drive Forest Park, GA 30297-2534 TECHNICAL REPORT STANDARD TITLE PAGE 1.Report No.: 2. Government Accession No.: 3. Recipient's Catalog No.: 4. Title and Subtitle: Life Cycle Cost Assessment And Performance Evaluation Of Sediment Control Technologies 5. Report Date: October 2015 6. Performing Organization Code: 7. Author(s): 8. Performing Organ. Report No.: S.E. Burns and C.F. Troxel (Georgia Institute 10-24 of Technology) 9. Performing Organization Name and Address: Georgia Institiute of Technology, School of Civil and Environmental Engineering 790 Atlantic Drive Atlanta GA 30332-0355 10. Work Unit No.: 11. Contract or Grant No.: 12. Sponsoring Agency Name and Address: Georgia Department of Transportation Office of Materials & Research 15 Kennedy Drive Forest Park, GA 30297-2534 13. Type of Report and Period Covered: Final; October 2010 October 2015 14. Sponsoring Agency Code: 15. Supplementary Notes: 16. Abstract: 17. Key Words: 18. Distribution Statement: 19. Security Classification (of this report): Unclassified 20. Security Classification (of this page): Unclassified 21. Number of Pages: 22. Price: Form DOT 1700.7 (8-69) GDOT Research Project No. RP 10-24 Final Report LIFE CYCLE COST ASSESSMENT PERFORMANCE EVALUATION OF SEDIMENT CONTROL TECHNOLOGIES By Susan E. Burns, Ph.D., P.E. Professor and Cameron F. Troxel, PE Graduate Research Assistant Georgia Institute of Technology Contract with Georgia Department of Transportation In cooperation with U.S. Department of Transportation Federal Highway Administration October, 2015 The contents of this report reflect the views of the author(s) who is (are) responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the Georgia Department of Transportation or the Federal Highway Administration. This report does not constitute a standard, specification, or regulation. TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. vi LIST OF FIGURES ......................................................................................................... viii List of Acronyms ............................................................................................................... xi ACKNOWLEDGEMENTS ............................................................................................. xiv EXECUTIVE SUMMARY ...............................................................................................xv INTRODUCTION ...............................................................................................................1 LITERATURE REVIEW ....................................................................................................2 Erosion and sediment control .......................................................................................... 2 Non-point source pollution .......................................................................................... 2 2.1.2 Erosion and sediment control regulation ............................................................ 5 2.1.3 Erosion prevention.............................................................................................. 6 2.1.4 Sediment control................................................................................................. 7 2.1.5 Temporary perimeter sediment control devices ................................................. 9 2.2 Performance of sediment control devices (SCDs) .................................................. 11 2.2.1 Comparison to geotechnical and environmental filters .................................... 11 2.2.2 Sedimentation of solids .................................................................................... 12 2.2.3 Filtration of solids............................................................................................. 15 2.2.4 Removal of dissolved contaminants ................................................................. 18 2.2.5 SCD selection ................................................................................................... 20 2.2.6 Measuring SCD performance ........................................................................... 21 2.2.7 Environmental impacts related to SRD manufacture and installation.............. 26 2.3 Life cycle assessment (LCA) .................................................................................. 28 2.3.1 Introduction to LCA ......................................................................................... 28 2.3.2 Performing an LCA .......................................................................................... 30 2.3.3 Goal and scope.................................................................................................. 31 2.3.4 Life cycle inventory (LCI)................................................................................ 33 2.3.5 Life cycle impact assessment (LCIA) .............................................................. 36 2.3.6 TRACI .............................................................................................................. 37 2.3.7 Valuation .......................................................................................................... 39 iii MATERIALS AND METHODS.......................................................................................41 3.1 Materials.................................................................................................................. 42 3.1.1 Sediment control devices (SCDs)..................................................................... 42 3.1.2 Test soil............................................................................................................. 47 3.2 SCD field testing ..................................................................................................... 48 3.2.1 ASTM D7351 test equipment........................................................................... 48 3.2.2 SCD installation................................................................................................ 50 3.2.3 SCD end connections........................................................................................ 53 3.2.4 ASTM D7351 test............................................................................................. 54 3.3 Laboratory analysis ................................................................................................. 54 3.3.1 Turbidity ........................................................................................................... 54 3.3.2 Total, suspended and dissolved solids .............................................................. 55 3.3.3 SCD nutrient retention...................................................................................... 55 3.3.4 Metal retention.................................................................................................. 58 3.4 Life cycle analysis (LCA) ....................................................................................... 59 3.4.1 Goal and scope.................................................................................................. 59 3.4.2 Functional unit (model conditions)................................................................... 59 3.4.3 System boundaries ............................................................................................ 60 3.4.4 Processes........................................................................................................... 60 3.4.5 Plans.................................................................................................................. 66 3.4.6 Impact analysis ................................................................................................. 74 RESULTS AND ANALYSIS............................................................................................75 4.1 SCD field performance............................................................................................ 75 4.1.1 Soil and water retention .................................................................................... 82 4.1.2 TSS reduction ................................................................................................... 87 4.1.3 SCD removal efficiency ................................................................................... 93 4.1.4 Turbidity reduction ........................................................................................... 95 4.1.5 Total dissolved solids (TDS) reduction .......................................................... 102 4.1.6 SCD installation.............................................................................................. 103 4.1.7 Discussion....................................................................................................... 104 4.2 SCD laboratory performance ................................................................................ 106 4.2.1 Nutrient retention............................................................................................ 108 4.2.2 Metal retention................................................................................................ 111 iv 4.2.3 Discussion....................................................................................................... 111 4.3 Life cycle analysis ................................................................................................. 112 4.3.1 Use-phase description.................................................................................... 112 4.3.2 Inventory balance ........................................................................................... 114 4.3.3 Life cycle impact analysis (TRACI).............................................................. 119 4.3.4 Discussion...................................................................................................... 126 4.3.5 Overall Performance (Valuation) ................................................................... 127 4.3.5 Implications .................................................................................................... 132 4.3.6 Limitations...................................................................................................... 134 CONCLUSION ................................................................................................................136 5.1 Sediment Control Device Performance.............................................................. 136 5.2 Life Cycle Analysis of Sediment Control Devices ............................................ 138 5.3 Impact/Recommendations .................................................................................. 139 5.4 Future Work ....................................................................................................... 140 APPENDIX A: DOT SUMMARY ...............................................................................141 APPENDIX B: TSS/TDS RESULTS ...........................................................................149 APPENDIX C: TURBIDITY RESULTS .....................................................................156 APPENDIX D: NUTRIENT RESULTS ......................................................................160 APPENDIX E: METAL RESULTS .............................................................................164 v LIST OF TABLES Table 1: Construction site runoff contaminants (from CALTRANS 2002). ...................... 4 Table 2: Filtration mechanisms (adapted from Ives 1973). .............................................. 17 Table 3: SCD performance from flume, sloping soil bed and cut trench testing.............. 24 Table 4: SCD performance from large scale testing. ........................................................ 25 Table 5: Developing a functional unit for cup comparison. ............................................. 32 Table 6: Impact categories considered in TRACI (Bare et al. 2003)................................ 37 Table 7: Common weighting factor development methods. ............................................. 41 Table 8: Description of sediment control devices tested .................................................. 43 Table 9: Material weight by unit length of device ............................................................ 43 Table 10: Description of field and laboratory test material .............................................. 44 Table 11: Soil gradation indicated in ASTM D7351. ....................................................... 50 Table 12: Nutrient test methods and description. ............................................................. 57 Table 13: Inputs and outputs from use-phase processes ................................................... 68 Table 14: Single-unit processes approximated for this research ...................................... 69 Table 15: Summary of ASTM D7351 testing................................................................... 75 Table 16: Range and average TSS for each test. .............................................................. 92 Table 17: Average upstream turbidity measurements. ..................................................... 99 Table 18: Available SCD installation equipment ........................................................... 104 Table 19: Schedule of laboratory testing ........................................................................ 106 Table 20: Percentage soil, nutrient or metal retained ..................................................... 112 Table 21: Solids, nutrients and metals (kg) leaving site over 1 year period. .................. 113 Table 22: Global warming potential (GWP) by process................................................. 120 Table 23: Acidification potential (AP) by process ......................................................... 121 Table 24: Weighting factors for LCIA valuation............................................................ 128 Table 25: Relative SCD performance, air quality........................................................... 129 Table 26: Relative SCD performance, water quality. ..................................................... 129 Table 27: Weighting factors for SCD performance valuation. ....................................... 131 vi Table 28: Relative SCD performance, TSS removal and turbidity reduction. ............... 131 Table 29: SCD TSS and turbidity removal performance. ............................................... 136 Table 30: Nutrient retention results. ............................................................................... 137 Table 31: Metal retention results. ................................................................................... 137 vii LIST OF FIGURES Figure 1: Source of impairment of Georgia rivers and streams (data from GAEPD 2010). ................................................................................................................................. 3 Figure 2: Site perimeter for a linear project. ....................................................................... 8 Figure 3: Sediment control devices used in states. ............................................................. 9 Figure 4: Bridging network at filter/soil interface . .......................................................... 12 Figure 5: Sedimentation behind barrier (adapted from McFalls 2009). ........................... 13 Figure 6: Settling velocity for 0.1, 1 and 5 mm diameter particles in transitional flow. .. 14 Figure 7: Settling velocity for 0.001 to 0.1 mm particle diameters in laminar flow. ....... 15 Figure 8: Surface charge on particle (adapted from Santamarina and Klein 2001) ......... 16 Figure 9: Dominant transport mechanism by particle size. .............................................. 18 Figure 10: Nitrate removal by denitrification (from Kim et al. 2003).............................. 20 Figure 11: Flume for testing SCDs (figure from ASTM D5141) ..................................... 23 Figure 12: Definition of life-cycle (USEPA 1993)........................................................... 28 Figure 13: LCA flowchart and description ....................................................................... 31 Figure 14: Unit versus system process (Processes adapted from www.madehow.com/Volume-3/Lumber). ............................................................ 34 Figure 15: Power grid mix in United States (PE INTERNATIONAL 2012)................... 35 Figure 16: Flow of work. .................................................................................................. 42 Figure 17: Mulch size gradation by sieve analysis. .......................................................... 45 Figure 18: Size gradation for two types of compost by sieve analysis ............................. 46 Figure 19: Determination of moist density. ...................................................................... 47 Figure 20: Test soil size gradation. ................................................................................... 48 Figure 21: ASTM D7351 testing equipment at DDRF. .................................................... 49 Figure 22: Approximate configuration of installed SCDs. ............................................... 53 Figure 23: Laboratory filter test configuration. ................................................................ 58 Figure 24: LCA system boundary..................................................................................... 60 Figure 25: Polypropylene fibers flow diagram (from GaBi 6.0). ..................................... 62 viii Figure 26: Timber process flow diagram (from GaBi 6.0) ............................................... 63 Figure 27: Municipal landfill processes (GaBi 6.0).......................................................... 66 Figure 28: Process plan for silt fence Type A................................................................. 70 Figure 29: Process plan for silt fence Type C................................................................ 71 Figure 30: Process plan for compost sock ........................................................................ 72 Figure 31: Process plan for straw bale barriers................................................................. 73 Figure 32: Process plan for mulch berm ........................................................................... 74 Figure 33: Type A silt fence during (left) and after (right) test. ....................................... 76 Figure 34: Type C silt fence during (left) and after (right) test. ....................................... 77 Figure 35: 12-inch compost sock during (left) and after (right) test................................. 78 Figure 36: Straw bales before (left) and after (right) test. ................................................ 79 Figure 37: Mulch berm before (left) and after (right) test. ............................................... 80 Figure 38: 18-inch compost sock before (left) and after (right) test................................. 81 Figure 39: Recorded tank weights and retention of Type A silt fence. ............................ 82 Figure 40: Recorded tank weights and retention of high-flow Type C silt fence with failure from undercutting at 18 minutes................................................................ 83 Figure 41: Recorded tank weights and retention of high-flow Type C silt fence............. 83 Figure 42: Recorded tank weights and retention of 12-inch compost sock with significant undercutting of beginning at 5 minutes. ............................................................... 84 Figure 43: Recorded tank weights and retention of 18-inch compost sock. ..................... 84 Figure 44: Recorded tank weights and retention of straw bales. ...................................... 85 Figure 45: Recorded tank weights and retention of mulch berm. ..................................... 85 Figure 46: Maximum exit rates for each SCD tested........................................................ 87 Figure 47: Measured TSS downstream of Type A silt fence, 5-minute samples. ............ 89 Figure 48: Measured TSS downstream of Type C silt fence, 5-minute samples. Fence failed 18 minutes into test. .................................................................................... 89 Figure 49: Measured TSS downstream of Type C silt fence retest, 5-minute samples. ... 90 Figure 50: Measured TSS downstream of 12-inch compost sock, 5-minute samples. ..... 90 Figure 51: Measured TSS downstream of 18-inch compost sock, 5-minute samples. ..... 91 Figure 52: Measured TSS downstream of straw bales, 5-minute samples. ...................... 91 Figure 53: Measured TSS downstream of mulch berm, 5-minute samples. ..................... 92 ix Figure 54: Cumulative solids passing SCDs..................................................................... 94 Figure 55: Removal efficiency including solids removed by sedimentation.................... 95 Figure 56: Measured turbidity downstream of Type A and Type C silt fence installations. ............................................................................................................................... 96 Figure 57: Measured turbidity downstream of 12 and 18 inch compost socks. ............... 97 Figure 58: Measured turbidity downstream of mulch berm and straw bales.................... 98 Figure 59: Measure turbidity of upstream samples........................................................... 99 Figure 60: Average turbidity reduction for entire test, first 30 minutes and time after first 30 minutes. .......................................................................................................... 100 Figure 61: Correlation for turbidity and TSS.................................................................. 101 Figure 62: Turbidity vs. TSS for straw bale and 12" compost sock. .............................. 102 Figure 63: Measured TDS for each SCD tested.............................................................. 103 Figure 64: Initial and replicate nutrient results, 0 2 mg/L ........................................... 107 Figure 65: Initial and replicate nutrient results, 0 0.5 mg/L ........................................ 107 Figure 66: Initial nutrient concentrations (mg/L) ........................................................... 108 Figure 67: Nutrient concentrations after ninth liter of water .......................................... 109 Figure 68: Nutrient retention of compost (upper), mulch (middle) and straw (lower). Negative is nutrient (mg/L) added to stream, positive is nutrient (mg/L) removed from stream. ........................................................................................................ 110 Figure 69: Metals retained in filter material ................................................................... 111 Figure 70: Soil loss balance ............................................................................................ 115 Figure 71: Energy resources by SCD.............................................................................. 115 Figure 72: Energy by production phase .......................................................................... 116 Figure 73: Non-renewable resources .............................................................................. 116 Figure 74: Freshwater emissions. Soil and water (runoff and process) not included ..... 117 Figure 75: Top 6 in-organic emissions to fresh water .................................................... 118 Figure 76: Top 5 emissions to air ................................................................................... 118 Figure 77: Global warming potential (GWP) ................................................................. 120 Figure 78: Acidification potential (AP) .......................................................................... 121 Figure 79: Eutrophication potential (EP) ........................................................................ 123 Figure 80: Ecotoxicity to water....................................................................................... 124 x Figure 81: GWP by phase ............................................................................................... 125 Figure 82: AP by phase................................................................................................... 125 Figure 83: Results of LCIA valuation. SCDs are ranked from lowest to highest relative environmental impact.......................................................................................... 130 Figure 84: Results of SCD performance valuation. SCDs are ranked from best (highest) to worst (lowest) relative performance. .............................................................. 132 Figure 85 ......................................................................................................................... 170 AHP AOS AP API ASTM BMPs CALTRANS CS CWA DDRF E&SC EI ELG EP List of Acronyms Analytical hierarchy process Apparent opening size Acidification potential Active pharmaceutical ingredients American Society for Testing and Materials Best management practices California Department of Transportation Compost sock Clean Water Act of 1972 Denver Downs Research Facility Erosion and sediment control Environmental Impact (EI) Effluent limitation guidelines Eutrophication potential xi ETP FU GAEPD GVWR GWP GSWCC ICP-OES IPCC ISO LCA LCI LCIA MUSLE NPDES NREL NTU NW PAF PAM PP PPA SETAC SF Ecological toxicity potential Functional unit Georgia Environmental Protection Division Gross vehicle weight rating Global warming potential Georgia Soil and Water Conservation Commission Inductively coupled plasma optical emission spectroscopy Intergovernmental Panel on Climate Change International Organization for Standardization Life cycle analysis Life cycle inventory Life cycle impact assessment Modified Universal Soil Loss Equation National Pollutant Discharge Elimination System National Renewable Energy Laboratory Nephelometric turbidity units Non-woven geotextile Potential affected fraction Polyacrylamide Polypropylene fibers Pollution Prevention Act of 1990 Society of Environmental Toxicology and Chemistry Silt fence xii SW Straw wattle SWPPP Storm water pollution prevention plans TDS Total dissolved solids TN Total nitrogen TP Total phosphorus TRACI Tool for the Reduction and Assessment of Chemical and other environmental Impacts TS Total solids TSS Total suspended solids UNEP United Nations Environmental Programme USEPA United States Environmental Protection Agency W Woven geotextile xiii ACKNOWLEDGEMENTS The authors are pleased to acknowledge the contributions of Mr. Zhi Ge, who worked as graduate research assistant on this project. In addition, the authors are very grateful to Mr. Darrell Richardson for his significant contributions in the formulation of this project and throughout the course of this work. Thanks for Mr. Jon D. Griffith for review of the final document. xiv EXECUTIVE SUMMARY This study was performed for the Georgia Department of Transportation (GDOT) to better understand the environmental impacts associated with sediment control technology currently employed on transportation projects. In this study, a review of current and past methods for testing sediment control devices (SCDs), both in the field and in the laboratory, as well as procedures for conducting a life cycle assessment was performed. Life cycle analysis (LCA) is discussed in depth in this report to facilitate future LCA on this subject. Field and laboratory testing is executed to measure performance of five different SCDs for retention of sediment, metals, and nutrients. Results of the tests are combined with existing data for the production and disposal of metal, plastic, and timber and emission data for trucks and machinery to model the life cycle of each SCD. An environmental impact analysis was performed using GaBi 6.0 software and USEPA TRACI methodology. Results of the impact analysis indicate: Straw bale installations significantly increase eutrophication potential in downstream water systems due to high levels of phosphate present in the straw bales. Production of steel sections and wire mesh for support of low permittivity Type C silt fence result in large increases in global warming and acidification potential. Performance of high permittivity Type A silt fence suggests that it is a good alternative to low permittivity silt fence in high volume and high sediment runoff conditions. xv The overall low global warming and acidification potentials of mulch berms, as well as their low aquatic toxicity levels, suggests that their use as an alternative to geotextile silt fence is favorable. xvi INTRODUCTION Sediment control is a critical aspect of any construction project. Contamination of waterways by suspended sediments, nutrients, and metals have clearly demonstrated detrimental impacts on the environment (Welsch 1991). High induced nutrient loads lead to eutrophication that can reduce infiltrating sunlight and result in a reduction in aquatic vegetation and animals. Transport of dissolved toxic chemicals or chemicals attached to suspended solids flowing in runoff waters increase toxicity in aquatic environments. Increased toxicity in surface water not only degrades the water environment, but it may lead to degradation in adjoining air and soil environments. Due to the substantial adverse impacts that result from runoff of solids to water ways, significant resources are invested in preventing erosion and retaining solids on land, with retention most commonly accomplished through the use of silt fences. Currently, silt fence is GDOT's predominant method for erosion control on construction projects in the early phases of construction, with GDOT installing approximately 1.0 1.5 million linear feet of silt fence per year. Silt fences are comprised of slit film, woven geotextiles, typically manufactured from polypropylene. Research and assessment of environmental impacts from silt fence largely focuses on impacts from effluent passing the silt fence installation. The focus on impacts due to effluent is likely because current regulations only describe impacts from construction site effluent and encourage minimal site runoff. While silt fence is a reliable form of sediment control, it impacts the environment in other ways; primarily due to construction from materials derived from fossil fuels, and due to the relative non-degradability of plastics in the long term. However, alternative erosion control technologies also exist, which rely on more environmentally friendly materials such as compost. This report is a presentation of the following research findings: 1 1. A comparison of the performance of multiple sediment control devices, 2. A performance a life cycle assessment of sediment control technologies (e.g., silt fence) that are currently in use by the Georgia Department of Transportation (GDOT) and, 3. A comparison of the performance and life cycle impacts for alternative technologies that are biodegradable and likely less dependent on fossil fuels for manufacture (e.g., straw bales, mulch berms, and compost filter socks) than plastic silt fence. Performance of sediment control devices (SCDs) is measured using a combination of largescale field testing and laboratory filter testing. Results of the performance tests are combined with material descriptions and used to generate `use' phases for each SCD. The use phases are combined with industry standard production information for polypropylene, steel, and timer to map the life cycle of each SCD for a model site. The life cycle maps, inventories, and impact assessment is performed using the GaBi 6.0 LCA software by PE INTERNATIONAL. Results of the life cycle impact analysis are modified using assumed impact category weighting factors. The results of the LCA are compared with typical SCD performance results to demonstrate the importance of life-cycle analysis. LITERATURE REVIEW Erosion and sediment control Non-point source pollution Recent assessment of waterways performed by the Georgia Environmental Protection Division (GAEPD) indicates the major contributor to impairment of rivers and streams is due to non-point source pollution (GAEPD 2010). Figure 1 shows the results of the river and stream 2 assessment from 2008 2009. Non-point source pollution can generally be described as a mix of water and contaminants that results from overland flow of rainfall and snowmelt (GAEPD 2010). Contamination of surface water by non-point source runoff waters has been proven to be harmful to aquatic life and vegetation (Welsch 1991). Suspended soil particles increase turbidity of waterways, thus blocking sunlight and debilitating growth of aquatic plants. Deposited sediment disrupts aquatic life at the bottom of stream beds and reduces flow areas, which increases the risk of flooding (Welsch 1991). Deposition of eroded sediments in waterways important for drainage and/or navigation may lead to costly dredging operations (Harbor, 1999). Transport of dissolved nutrients or nutrients attached to sediments cause algae blooms that limit light penetration below the water surface, preventing photosynthesis in lower aquatic plants and lowering dissolved oxygen concentrations. Miles of river and streams in Georgia impaired by source 5841 2189 4 57 93 179 274 Figure 1: Source of impairment of Georgia rivers and streams (data from GAEPD 2010). 3 Suspended solids concentrations in runoff water from cleared farmland can be as high as 20,000 mg/L (Carter et al. 1993). Erosion occurring on active construction sites is amplified by clearing and grubbing activities, soil stockpiling and general earth moving. McCaleb and McLaughlin (2008) report turbidity and total suspended solids (TSS) concentrations leaving five sediment traps at construction sites in the Piedmont region of North Carolina as varying from a minimum of 0 to over 30,000 nephelometric turbidity units (NTU) and from 2 to greater than 168,000 mg/L, respectively. Construction sites are also a source for environmentally harmful chemicals and heavy metals from machinery emissions and leachate from concrete or other pavements. High nutrient loadings in runoff water during time of grassing and seeding slopes are also possible. A study by the California Department of Transportation (CALTRANS) was conducted from 1998 to 2002 in order to characterize runoff from transportation construction sites (CALTRANS 2002). Minimum, maximum and mean values reported through the study are shown in Table 1. The large range in contaminant concentrations is due to the variability in construction processes and in impacts associated with each process. Table 1: Construction site runoff contaminants (from CALTRANS 2002). Contaminant TSS TDS Turbidity Nitrate Nitrite Total Phosphorous Ammonia Unit mg/L mg/L NTU mg/L mg/L mg/L mg/L Min Max Avg 12 3,850 472 22 1,270 225 15 16,000 636 0.12 3.90 0.95 0.10 2.80 0.16 0.05 15.00 1.02 0.06 4.00 0.29 4 Copper, Dissolved g/L Lead, Dissolved g/L Zinc, Dissolved g/L 1.00 29.80 7.29 0.50 36.50 1.11 1.00 209.00 17.50 2.1.2 Erosion and sediment control regulation Federal and state (Georgia) legislation pertinent to erosion and sediment control regulations is summarized in the following paragraphs. Unless otherwise noted, descriptions are according to the United Stated Environmental Protection Agency (USEPA 2002). Clean Water Act of 1972 (CWA) Prohibited unauthorized release of pollutants into navigable and other connected waterways. Established the National Pollutant Discharge Elimination System (NPDES) which requires permits for all point-source discharges to significant waterways. Required USEPA to develop effluent limitation guidelines (ELGs) for major point source categories. Georgia Erosion and Sedimentation Act of 1975 (Act 599) Required counties and municipalities of Georgia to develop plans to reduce releases of sediment from landdisturbing activities on sites 1 acre or larger. Plans developed are reviewed and approved by Soil and Water Conservation District. Plans must include permit process completed before land-disturbing activities commence (GSWCC 2000). Water Quality Act of 1987 Defined municipal and industrial storm water discharges as point sources thereby including them in NPDES regulation. o NPDES Phase I (promulgated 1990) Required construction sites of 5 or more acres to have individual or general discharge permit for storm water discharges. Permit requires preparation of storm water pollution prevention plans (SWPPP) that use best management practices (BMPs) as defined by state or local authorities. o NPDES Phase II (promulgated 1999) Extended most requirements of Phase I to small construction sites between1 to 5 acres. Pollution Prevention Act of 1990 (PPA) Prioritizes pollution prevention and environmentally safe pollution disposal. Related to site stabilization, erosion control and 5 proper maintenance as prevention measures versus sediment barriers as response measures. The most recent NPDES general permit for discharges from construction sites (NPDES 2012) requires that sites contain natural buffers or equivalent sediment controls when surface waters are located within 50 feet of the project's earth disturbances. Linear construction projects (e.g. highways, pipelines, sewers) are exempt from the requirement provided they limit disturbance, or provide supplemental sediment controls, to treat runoff within 50 feet of the surface water. 2.1.3 Erosion prevention Two modes of sediment management exist on construction sites: erosion prevention and sediment control. Erosion prevention is an attempt to stabilize slopes, disturbed areas and other surfaces susceptible to erosion and prevent detachment of soil particles from the ground surface. Sediment control is the capture and containment of eroded sediment and other pollutants being transported in runoff water. Erosion and sediment control (E&SC) represents a combined offensive (erosion prevention) and defensive (sediment control) approach to site runoff (Theisen 1992). Various materials are used for erosion prevention. Mulch applied as thin as 0.75 inches was shown to reduce soil erosion in sloping erosion test beds (Demars et al. 2004). Laboratory scale test results on erosion control products indicate erosion control performance to be closely related to the geotextile induced roughness, water-holding capacity and 24-hour wet weight (Rickson 2006). Geotextile is common in erosion control applications; one of the first major applications of geosynthetics was by the U.S. Army Corps of Engineers when they used a plastic sheet to stabilize soil under concrete blocks set to armor shorelines (Theisen 1992). Other erosion prevention practices are establishing vegetation (seed or sod), polyacrylamide (PAM), and soil tackifier and binders (GSWCC 2001). 6 Sediment control, not erosion prevention, is the topic of this study and erosion prevention is not discussed in this paper beyond this section. It is briefly described here primarily because erosion prevention devices are an important part of any E&SC plan when the construction site has exposed soil slopes. Also, testing of erosion prevention and sediment control devices is performed with similar methods but erosion products show more consistent results between studies. This may be due to erosion products being less installation dependent and/or more dependent on geotextile index properties. In contrast, performance of sediment control devices is very installation-dependent and often not dependent on material index tests. This topic is reviewed in detail in the next section. 2.1.4 Sediment control Sediment control is the effort to contain sediment in motion on site. Sediment controls operate by trapping sediment-laden runoff water and reducing sediment loads by sedimentation or filtering. Sediment control devices that are installed along the outer edge of sites to prevent sediment from escaping the site are termed perimeter controls or barriers. Other devices that are installed in concentrated flow conditions to slow the rate of runoff flow and to filter suspended solids are termed checks or dams. Perimeter control devices represent a significant portion of E&SC measures installed on highway construction and other linear projects because site perimeters are much larger for linear construction sites than for approximate square sites, as shown in Figure 2. 7 Figure 2: Site perimeter for a linear project. Many different perimeter sediment control devices are currently being used by state DOTs. An online review of current DOT perimeter controls was performed by visiting state DOT websites and searching for erosion and sediment control best management practices (BMPs) related manuals or guidance documents. All but one state provided DOT-specific guidance on E&SC BMPs or provided links to the governing state procedures. No searchable guidelines or links were found for Oklahoma. The results of the search are summarized in Figure 3. Each of the forty-nine states investigated include geotextile silt fence as a temporary sediment control best management practice (BMP). Twenty-two states list straw or hay bales as perimeter controls, three states do not allow bale installations. Earth berms and wattles are common. Less common perimeter controls are filter socks, brush barriers and vegetated buffer strips. Seven sites list a triangular filter dike as a perimeter control. The triangular filter dike is a 8 long three-sided wire (typical) mesh prism wrapped in geotextile fabric. The triangular dikes ranged in height from 8 to 18 inches. The dikes were transportable, reusable, did not require trenching and appeared to be an innovative approach to perimeter control. Silt fence Straw or hay bales Earth berm or diversion ditch Wattles - Straw, coconut,... Gravel/rock berm or filter sock Compost berm or filter sock Brush barrier Vegetated strip Mulch berm or filter sock Sediment trap Sand bag barrier Triangular silt dike 12 12 12 12 8 8 7 7 49 22 21 19 Number of time device was listed in E&SC manuals. Forty nine states investigated. Figure 3: Sediment control devices used in states. 2.1.5 Temporary perimeter sediment control devices The following represent the five most commonly referenced perimeter sediment control BMPs recommended by state DOTs or other responsible E&SC authority (i.e. silt fence, bales, berms, wattles and socks). More information, figures, and installation procedures can be found in the provided reference or references listed in Appendix A, or in the USEPA national menu of stormwater BMPs at: http://cfpub.epa.gov/npdes/stormwater/menuofbmps/. Geotextile silt fence (filter fence) is the most commonly installed perimeter control device seen on highway projects. Silt fence is typically woven polypropylene geotextile supported vertically by wooden or metal stakes. Some filtration occurs with new installations; however, geotextiles clog rapidly and the primary mode of sediment removal becomes flow retardation and subsequent sedimentation of suspended solids (Barrett et al. 1998, USEPA 2012). 9 Silt fence requires trenching, burial, and soil compaction and is easily damaged so performance is very installation dependent (Barrett et al. 1998, Zech et al. 2008) Straw or hay bales remove solids by retention and sedimentation. Individual bales have very low permeability but linear installations of bales are susceptible to leaks forming at the bale connections. Bales readily degrade in the field and may require replacement in as little as three months (USEPA 2010). The USEPA and many state entities do not recommend using bales for sediment control. USEPA (2010) recommends silt fence as a perimeter control alternative. Berms are linear piles of material used to retain and filter sediment-laden runoff water. Berms can be made of soil (also referenced as "diversion dike"), stone, mulch or compost. Minimum widths at the bottom of the berm and minimum berm heights are normally required and will vary depending on material Type and location. Soil, mulch and compost berms may require some compaction. Larger berms are less susceptible to erosion, offer greater filtration capabilities and are less likely to clog. For better confinement of materials, berms may be covered with geotextile (TxDOT 2004). Berms formed from brush debris created during site clearing (slash) are termed brush barriers. Brush barriers are typically installed on large sites that have been cleared and have sufficient working room at the perimeter for large berm installation. Filter socks are essentially a contained berm. Filter socks are often larger than wattles (subsequently discussed) and are normally filled with compost, mulch, or stone. With proper equipment for heavy lifting, installation of filter socks is not difficult. Performance depends on filter material and connection with the ground surface. Removal of sediment as well as pollutants (i.e. nutrients, coliform, e. coli, hydrocarbons) is possible with compost filter socks (Faucette et al. 2009). 10 Wattles (fiber rolls) are usually smaller in diameter than filter socks. Wattles are typically less than 12-inches in diameter and are filled with straw, coconut or wood fibers and wrapped with polypropylene netting, burlap, jute or coir. The best application of wattles is on contours along slopes to minimize sheet flow velocity and reduce surface erosion (EPA 2012). Wattles should be partially buried and staked. 2.2 Performance of sediment control devices (SCDs) 2.2.1 Comparison to geotechnical and environmental filters The performance of SCDs is directly related to their ability to filter sediment from runoff water. Soil filters have many applications in geotechnical and environmental engineering. Landfill leachate collection, soil stabilization, wall drainage and subsurface drains are some uses. Typically, these filters are designed to optimize soil retention and fluid discharge (Giroud 2006, Bhatia and Huang 1995). The first goal implies containment while the latter implies permeability and with good design, the proper balance is achieved. Filter criteria dictate that the filter media is selected to retain the largest soil particles to be filtered (Terzaghi and Peck 1967). Proper function of filters requires the development of a bridging pattern within the retained soil adjacent to the filter material (Bhatia and Huang 1995, Hoare 1982, Hongo and Veneziano 1989). If successful, the filter retains the largest soil particles, which in turn retain smaller and smaller particles. After some initial fall out, the retained soil develops a stable network with no additional release of soil particles. A stable bridging network and successful filter is shown in Figure 4. 11 Figure 4: Bridging network at filter/soil interface . The function of filters for sediment control is significantly different from that of filters in other geotechnical applications. The difference is because the formation of a stable filter layer is not possible, which is true for two reasons. First, eroded soil approaching sediment control devices is predominantly comprised of fine particles (Barrett et al. 1998) that pass through the filter unabated or clog filter pores. Second, coarse soil particles that have eroded will likely settle and deposit before reaching the sediment barrier. Many SCDs reduce sediment loads by retaining sediment-laden water and allowing for sedimentation of suspended solids. In this way, SCDs resemble Type I sedimentation basins, common during the initial steps of water treatment. Type I sedimentation refers to the settling of individual particles due to gravity forces. Type II and Type III sedimentation refers to flocculated and zone settling, respectively (Droste 2005). When SCDs are installed across contours, runoff water will flow behind and along the barrier (Beighley and Valdes 2009). This flow will induce turbulence or shear forces that may also remove filter cakes. 2.2.2 Sedimentation of solids Sedimentation is the process of removing solids from runoff flow long enough that suspended particles will be pulled down, out of suspension due to gravity forces. Figure 5 shows 12 sedimentation occurring behind a vertical SCD. The rate of mass settling behind the barrier is found with a simple mass balance: = 11 - 22 Where dm/dt is the positive change in mass settling behind the barrier, Q1 and C1 are the influent flow rate and concentration and Q2 and C2 are the effluent flow rate and concentration, respectively. Units of flow rate are volume per time; units of concentration are mass per volume. V Q2,C2 Q1,C1 Figure 5: Sedimentation behind barrier (adapted from McFalls 2009). Settling particles can be sorted into two sizes, (1) large particles with settling velocities much higher than the velocity of water draining through the filter and (2) small particles that settle after a long detention time behind barriers. Settling velocity for spherical particles in water is calculated with the following equation: = 4 3 - Where v = settling velocity (L/T), g = gravitational constant, d = diameter of sphere, CD = drag coefficient, s = density of solid particle, and f = density of the fluid. The drag coefficient varies 13 according to the Reynold's number (Re) for laminar and transitional flow according to the following equations: Laminar flow (Re < 1.0) 24 = Transitional flow (1 98% pure) propylene and accounts for one half of the world consumption of propylene (Aitani 2006). The majority of propylene is created in parallel processes during the steam cracking of naphtha or gas oil for ethylene and gasoline production. A high amount of energy and water as steam and quenching fluid is required for steam cracking reactions (Aitani 2006). Timber production includes various processes (i.e. debarking, sawing) with power requirements. Timber also requires a large amount of land usage (GaBi 6.0). Production of steel, plastic and timber require a high amount of diesel 26 operated trucks for transport. Operation of diesel fueled cargo trucks increase emissions with potential for global warming. 27 2.3 Life cycle assessment (LCA) 2.3.1 Introduction to LCA Life-cycle assessment or analysis (LCA) is a procedure used to measure the environmental impact of a product over the entire life of the product (SAIC 2006). The analysis is performed by summing the inputs and outputs for product phases that include raw-material extraction, production, distribution, use, and disposal or recycling and determining the effects of the cumulative inventory on the environment. LCA is also referred to as a cradle-to-grave assessment because it sums impacts from the product's creation (cradle) to the products disposal (grave). Figure 12 shows a typical flow through life-cycle stages and inputs used and outputs required by the entire system. Inputs Life-Cycle Stages Outputs Raw Materials Energy Raw Materials Acquisition Manufacturing Use/Reuse/Maintenance Recycle/Waste Management System Boundary Atmospheric Emissions Waterborne Wastes Solid Wastes Coproducts Other Releases Figure 12: Definition of life-cycle (USEPA 1993). 28 LCA has become an important feature in environmental analyses of industrial processes because of its ability to describe the interaction of large systems and their environments (Curran 1996). The holistic approach used by LCA is a popular method of environmental analysis in development of regulatory guidelines, product research and development, and as a rating tool used for product promotion (PR 2010). LCA is used to identify environmental impacts that are not apparent. Cook et al. (2012) used LCA to evaluate disposal options for active pharmaceutical ingredients (APIs). The increase in APIs encountered in water sources and the resulting potential environmental impacts has instigated API take-back or incineration programs, however; APIs are commonly discarded in municipal solid waste or flushed in drains. The analysis compared trash, toilet and take-back disposal options. Results show that while take-back programs remove APIs from the environment entirely, non-API emissions increase by more than 200% and global warming emissions increase by 1700% over baseline values. Internationally, the following three organizations perform research, develop guidelines and promulgate information on the useful application of LCA; (1) Society of Environmental Toxicology and Chemistry (SETAC), (2) International Organization for Standardization (ISO), and (3) United Nations Environmental Programme (UNEP). SETAC is a professional organization with members from academia, industry and government that assesses scientific research focused to improve the environment (Guine 2002). ISO provides worldwide standardization for many activities. ISO's biggest addition to LCA is the 14000 series of standards focused on environmental management systems. Included in the series is standard 14040, a framework for completing LCA that has been accepted worldwide (Guine 2002). UNEP generally is concerned with global application of LCA. 29 Nationally, LCA research has been performed by the United States Environmental Protection Agency (USEPA), the National Renewable Energy Laboratory (NREL), and the American Society for Testing and Materials (ASTM). ASTM has developed multiple standards for analysis of construction materials. The USEPA has developed a large amount of information pertaining to environmental LCA. The USEPA links to a wide range of LCA resources online at http://www.epa.gov/nrmrl/std/lca/resources.html. 2.3.2 Performing an LCA Life-cycle analysis is generally performed in four connected processes: description of goal and scope, life cycle inventory (LCI), life cycle impact assessment (LCIA), and interpretation. The four processes are described in Figure 13. Development and standardization of the LCA process began at an LCA workshop sponsored by SETAC in 1990 (Curran 1996). However, increased availability of process databases and processing capabilities of computers have allowed for added sophistication to LCA (Bare et al 2003). A discussion of each of the four LCA steps follow: 30 Goal & Scope Clearly define purpose of LCA and boundaries of system to be analyzed. List the impact categories to be included in the analysis. Life Cycle Inventory (LCI) Prepare flow diagram of energy and materials consumed during product phases. Sum input and output values for each phase of the product life. Life Cycle Impact Assessment (LCIA) Determine environmental cost of each product phase and for the product lifespan as a result of input and output values calculated during LCI. Interpretation Identify key issues indicated by LCA. Evaluate the reliability of study and determine sensitivity of data. Develop conclusions and limitations of the study. Figure 13: LCA flowchart and description 2.3.3 Goal and scope The initial step of performing an LCA is defining the goals of the study and outlining the project scope. The goal of the study should identify the specific purpose of the study, as well as the intended audience. The scope of the study should include the best definition of the functional unit to be compared, clearly state the system boundaries and define the limits for including inputs and outputs (PR 2010). When two items are compared, a functional unit (FU) is the best description of a task to normalize variations between the items. For example, consider three cups with characteristics defined in Table 5. To compare each cup a FU must account for volume and lifetime differences. Based on volume differences, 1.5 of cups B and C are required for comparison to cup A. Based on lifetime differences, more of cups A and B are needed for comparison with cup C. For the cup example, the impact of washing the reusable cup should be evaluated. 31 Table 5: Developing a functional unit for cup comparison. Volume (ounces) Uses per lifetime Functional unit (FU) Cups required per FU Cup A (disposable) 24 Cup B (disposable) 16 Cup C (reusable) 16 1 1 50 50 uses at 24 ounces per use 50 75 1.5 System boundaries are used to outline the extent of data to be analyzed in the study. Three orders of analysis are normally considered in LCA analysis: 1st order only production of materials and transport included, 2nd order all life cycle processes included, but no capital goods and 3rd order same as 2nd order with capital goods included (PR 2010). Capital goods are typically considered as equipment used for manufacture or installation. For LCA of complex systems with specialty manufacturing equipment, inclusion of the items is important. For comparison of products that use common procedures or machines, capital goods are normally not included. LCA is performed by developing a model of complex systems; inevitably, the model will be a simple version of the real system (PR 2010, Bare et al. 2003). A well defined goal is important to be able to properly determine the project's scope. For example, the scope of a study aimed to compare two interchangeable steps in a large manufacturing train may be much narrower than for a study to outline two different manufacturing trains. 32 2.3.4 Life cycle inventory (LCI) Compiling an LCI is the second step of performing an LCA and includes defining the process chain that occurs within the system boundaries and determining the inputs and outputs required/generated for each process. Requirements for each process are added together to determine the overall inputs and outputs required and created by an entire system. The resulting inputs and outputs for the entire system are shown in Figure 12. Creating a descriptive process chain for a complex system is difficult because it requires collaboration between groups with knowledge of systems occurring within a system. Overall, the LCI phase is the most dataintensive step of an LCA and requires diligent data management (PR 2010). Two types of processes are used to define systems, unit processes and system processes. A unit process is the smallest divisible function in a chain of processes. System processes represent a large combination of processes grouped together to describe a system occurring within a system (PR 2010). System processes are also known as cradle-to-gate or gate-to-gate processes because they represent multiple steps to get from one point to another (GaBi 6.0). An example of a chain of unit processes lumped into a system process for lumber manufacturing is shown in Figure 14. 33 Chain of Unit Processes Simplified Chain Access Road Construction Lumber Manufacturing Felling Transport Unloading Debarking/Bucking System Process Single Unit Process Headrig Sawing Bandsawing Drying Planing Grade Stamping Banding Transport Transport Stocking Stocking Figure 14: Unit versus system process (Processes adapted from www.madehow.com/Volume-3/Lumber). Developing a comprehensive LCI is simplified by the use of background data (PR 2010). Background data is information compiled and made available by the LCA industry that describes common processes. Background data are available in the form of LCI databases and usually come as part of any LCA software. The ecoinvent database by the Swiss Centre for Life Cycle Inventories ecoinvent is an extensive database with over 4,000 LCI datasets that cover multiple industries. The GaBi professional database has over 2,000 datasets that include metals, plastics, wood products, power generation and transport. Numerous other general or spatial and 34 industry-specific databases are available commercially. The databases are combinations of data developed by industry partners, government and private research that are reduced into a single format for easier application. Up to 80% of some LCIs can be completed by using information available in databases. When using background data, providing a process that adequately resembles the process it is going to represent is important (PR 2010). Power generation varies on the local level and represents a significant part of most life cycle assessments so power processes are usually developed using region-specific information. The historic distribution of power sources for the United States is shown in Figure 15. Figure 15: Power grid mix in United States (PE INTERNATIONAL 2012) Not all data required to complete an LCI will be available via an LCI database. Data that are not background data are termed foreground data and must be developed for the LCA study being performed. Foreground data are required for less common manufacturing processes, proprietary processes or items and processes performed on a local scale (GaBi 6.0). Developing 35 foreground data can be difficult. The most common sources of foreground data are industry surveys. Survey data requires a good understanding of the qualifications for persons providing survey information (PR 2010). Results of the LCI can be used to compare two processes. This type of comparison is called loading (Curran 1996). For example, if it is found that one process produces 2 metric tons of CO2 and another process produces 3 metric tons of CO2 per year then the first process is environmentally favorable. Comparisons using loading have many shortcomings. If two processes emit different levels of two different gasses it is not clear which process is better without knowing the environmental impacts of each gas (Curran 1996). 2.3.5 Life cycle impact assessment (LCIA) During LCIA, the environmental impacts associated with the LCI are assessed. LCIA uses results from the inventory phase as inputs into impact assessment metrics previously developed by the scientific community (Bare et al. 2003). Impact assessments are based on the best scientific knowledge of how material and energy uses and emissions effect the environment. LCIA is performed in the following order: 1. Compile stressors from LCI for process or item to be compared, 2. Classify stressors to impact categories, 3. Characterize stressors as equivalent units for each category (PR 2010). The compiling step aggregates inputs and outputs from each process in the life cycle inventory. Classification sorts emissions into impact categories. Inventory items can broadly be divided into two impact types; (1) depletions of raw materials by system inputs, and (2) pollutant emissions by system outputs. Inventories are further divided by potential damage to specific environmental or human health categories such as global warming potential, eutrophication 36 potential, or human cancer risk. Different emissions have different potential for impact. For comparison, emissions are usually converted to equivalent values of a standard unit. In the case of global warming, emissions are expressed as equivalent kg of CO2. The last part of the impact analysis is presenting the results of the LCA, usually in a graphical or tabular format. Characterizing the potential impact of material extraction or pollutant emission on a general impact area is considered characterization at the midpoint level. Global warming potential is a midpoint impact level. Sea-level rise and soil moisture loss are two of many possible results associated with global warming. Characterization at this level is considered endpoint characterization. At this time, there is little scientific consensus on how to properly quantify of emission impacts at end-point levels (Bare et al. 2003). 2.3.6 TRACI The impact assessment methodology used for this research is the Tool for the Reduction and Assessment of Chemical and other environmental Impacts (TRACI) developed by the USEPA. TRACI was created to provide a detailed impact analysis tool for use in the United States (Bare et al. 2003). TRACI quantifies impacts at midpoint levels only. Impact categories considered in TRACI are listed in Table 6. Table 6: Impact categories considered in TRACI (Bare et al. 2003) Ozone depletion Acidification Photochemical smog Human health: air pollutants Human health: non-cancer Land use Climate change Eutrophication Ecotoxicity Human health: cancer Fossil fuel depletion Water use 37 For this research, impact categories of climate change, acidification, eutrophication and ecotoxicity to water are considered; each category is described below. Climate change is the warming of the planet with possible endpoint effects including drought, floods, sea-level rise, loss of polar ice caps and change in weather patterns (Bare et al. 2003). Climate change is described in terms of global warming potential. Global warming potential (GWP) is presented in terms of equivalent mass of CO2. The Intergovernmental Panel on Climate Change (IPCC) is a Nobel Prize winning panel that provides most information on climate change. IPCC (2007) provides a list of emissions and GWPs for 20, 100 and 500 year periods. Degradation of pollutants in the atmosphere occurs at different rates and so GWP varies differently over time for different pollutants. Some notable emissions listed [and the associated 100-yr global warming potential] are carbon dioxide (CO2) [1], nitrous oxide (N2O) [298], methane (CH4) [25], Chloroflourocarbons (CFCs) [4,750 14,400] and hydrochloroflourocarbons (HCFCs) [124 14,800]. Cumulative GWP for products and processes is reported as the summation of each emission (kg) multiplied by its individual GWP value. Acidification is the increase in acidity of soil and water systems generally by acid rain (Bare et al. 2003). Acidification endpoint effects are reduction in lake alkalinity, corrosion of buildings and other structures and plant and animal death. Acidification potential (AP) is described in terms of equivalent kg of hydrogen (H+) ions. Major contributors to AP are sulfur dioxide (SO2), nitrogen oxides (NOX), strong acids and ammonia (NH4). Unlike GWP, AP depends on deposition characteristics of emissions in the local environment, therefore; cumulative AP for products and processes is the summation of each emission converted to equivalent kg H+ ions and multiplied by an environmental deposition factor (Bare et al. 2003). 38 Deposition factors describe the amount of H+ ions expected from Emissions of SOx and NOx. The deposition factors are determined by atmospheric chemistry and transport equations and vary for different regions within the United States. Eutrophication is the release of nutrients to the environment at levels much higher than normal. The addition of fertilizing nutrients increases plant life (algae) at the water surface that leads to reduction of sunlight infiltration into the water body and reduction of oxygen levels (SAIC 2006). Eutrophication potential (EP) is described as an equivalent mass of nitrogen (N). Nitrogen and phosphorous (P) emissions are responsible for the majority of EP. The effect of N and P releases into the environment is dependent on existing N and P levels in the local environment. Cumulative EP for products and processes is the summation of each emission converted to equivalent nitrogen (kg) and multiplied by two environmental impact factors, one for transport capability and one for existing nutrient levels (Bare et al. 2003). Ecotoxicity is a measure used to quantify possible damages due to discharging toxic materials into the soil, water and air environments. For harmful contaminants, an ecological toxicity potential (ETP) is developed that describes the potential impact for that contaminant when released into the environment (Bare et al. 2003). Ecotoxicity is reported for water, air and soil but each is related; emissions to soil may lead to damage in the air and water and vice-versa. Cumulative ecotoxicity for products and processes is the summation of each emission multiplied by a cumulative ETP and reduced for transportation and degradation characteristic of the individual contaminant (Bare et al. 2003). 2.3.7 Valuation At times, a best-performing alternative can be identified from the unmodified results of the LCIA. The best-performing option performs better in every impact category analyzed, or 39 significantly better in one or more categories and equal to the alternatives in the remaining categories. When more than one option performs significantly better than other options in different impact categories, a best option cannot be determined without comparing the significance of the impact categories. Applying weighting factors to impact categories based on their supposed importance to facilitate comparison is referred to as valuation or weighting (SAIC 2006). Using weighting factors, a single environmental impact can be calculated for each process or unit to be compared. Calculating the total environmental impact is expressed by the following equation from Finnveden (1999). = =1 Where EI is the total environmental impact from n impacts, V is the value weighting factor for impact category i, and I is the impact from one unit in category i. Multiple problems arise when developing weighting factors. First, no consensus for relative importance of impact categories has been determined. Therefore, weighting factors are subjective values selected based on opinion and will vary from person to person. Second, the relative importance of impact factors may change with location and time (SAIC 2006). Although there are obvious shortcomings, thoughtfully developed weighting factors will greatly aid the decision-making process. Three commonly used systems to develop weighting factors are panels, distance to target, and monetization (Pr 2010). Detailed descriptions and comparison of other available weighting methods are discussed in Finnveden (1999). Weighting factors using the monetization method are developed and discussed in Johansson (1999). Berrittella et al. (2007) developed weighting factors to aid in selecting transport policies to reduce climate change impacts using the analytical hierarchy process (AHP) and pair-by-pair comparisons. 40 Table 7: Common weighting factor development methods. Panels Distance to Target Monetarisation A panel is developed with industry, environmental, and political professionals. Weighting factors are either voted on or developed using a multi-criteria analysis and comparison by pairs. Weighting factors are determined based on the distance to reduction target (determined by environmental policy). Environmental impacts are expressed as a monetary cost required for clean-up. Higher cost is weighted higher. No investigation was performed to determine the best weighting factors to be used for interpretation of LCIA results in this study. Determining weighting factors will require discussion between GDOT, environmental experts, and the USEPA. To demonstrate the benefits of an LCIA valuation process, this study includes a simple value calculation with assumed weighting factors. The weighting factor assumptions, LCIA valuation, and areas of improvement are discussed in section Error! Reference source not found. of this report. MATERIALS AND METHODS A combination of field and laboratory testing was performed to characterize SCD performance during the utilization (use) phase. Performance was based on suspended solid, nutrient, and metal retention capability. Material production and approximated installation and maintenance processes were combined with the use-phase and used to compile an inventory of materials and emissions. Life cycle impact analysis was performed using GaBi 6.0. The flow of work is described in Figure 16. 41 Life cycle impact analysis 3rd (LCIA) to compare SCD environmental impacts 2nd Use phase processes Installation and maintenance processes Material production processes 1st Field testing for sediment retention Lab tests for nutrient and metal retention Figure 16: Flow of work. 3.1 Materials 3.1.1 Sediment control devices (SCDs) Large-scale field testing of SCD performance testing were performed on five different commonly used SCDs. The five devices were: 1. Silt fence - Type A, 2. High-flow silt fence - Type C, 3. Compost sock, 4. Straw bale, and 5. Mulch berms. Silt fence materials were purchased from a construction supply warehouse, compost socks were provided and delivered to the test facility by Filtrexx, straw bales were available at the test site, mulch was purchased from a local bulk supplier in Anderson, SC. Initial testing of a 12" diameter compost sock resulted in significant undermining of the sock. A retest was performed using a larger diameter (18 inch) sock. Table 8 lists the source and provides a short description of each material tested. Table 9 shows the measured distribution of material type per unit (meter) 42 length of SCD device. Lengths of polypropylene geotextile and steel wire mesh were cut and measured for length and weight. Wood and steel stakes were weighed individually and divided over the installation increment (4' for Type C, 6' for Type A and compost sock, 2 per straw bale). An approximate moist density of installed compost and mulch samples was determined using laboratory index tests and the equivalent weight per length value calculated by multiplying by the installed device volume for one unit length. Table 8: Description of sediment control devices tested Material Silt fence (Type A) High-flow silt fence (Type C) Type ErosionTech ET-GA-A ErosionTech ET-GA-C Description Polypropylene monofilament woven fabric with wood fence stakes Polypropylene monofilament heat bonded fabric with 4 inch square steel wire mesh and steel fence stakes Compost sock 12" Filtrexx silt sock 18" Filtrexx silt sock Polypropylene mesh sock with compost fill Straw bale Available at test site Typical 36" straw bale Mulch berm Coastal Bark & Supply, local Unpainted , unspecified mixed hardwood mulch, typical mulch supplier for landscaping Table 9: Material weight by unit length of device Device Material Type A fence Type C fence Compost sock PP Wood stakes PP Metal stakes Metal wire mesh PP Measured [calculated] weight (g) 331.570 779.380 546.570 2117.170 989.390 170.190 Length increment (m) 3.048 1.829 3.048 1.219 3.048 1.029 Weight per length (kg/m) 0.109 0.426 0.179 1.737 0.325 0.165 43 Straw bale Mulch berm Compost Wood stakes Straw Wood stakes mulch [103600] 1385.560 14288.148 1385.560 [59400] 0.305 1.829 0.914 0.457 0.305 103.600 0.758 15.626 3.031 59.400 Laboratory filtration tests were performed using samples of compost and straw from the field SCD devices and samples of two different types of mulch. Mulch samples were provided by a local Atlanta bulk supplier. Mulch samples were a cypress wood waste from southern Georgia and a mixed bark, cambium, and hardwood mulch. A description of each material tested is included in Table 10. Size gradation analysis was performed on samples of mulch and compost from each compost sock used. Gradation was performed by separation through a stack of eight sieves ranging in opening size from 76.2 mm (3 in.) to 0.81 mm (#20 sieve). Gradation of the mulch is shown in Figure 17. Gradation of both compost samples is shown in Figure 18. Table 10: Description of field and laboratory test material Sample Compost 1 Compost 2 Field mulch Cypress mulch Source 18" compost sock 12" compost sock Mulch berm Atlanta supplier Description Dark brown to black to 3-inch long rounded wood pieces with a significant amount of organic soil fines, some fine to medium crushed stone and plastic waste Light brown to brown to 5-inch long slender wood mulch with a significant amount of reddish brown fine soil Dark brown to 3-inch long chopped wood fragments, some fine wood particles Moisture Content (%) 67 75 177 Yellowish brown to 2-inch long chopped wood fragments, small amount of fine particles 44 Mixed hardwood mulch Atlanta supplier Dark reddish brown mix of 1/8 to 2-inch long broken round wood particles, to 5-inch long wood cambium strips and to -inch wood bark pieces Straw Straw bale installation Standard yellow straw < 10 PERCENT FINER BY WEIGHT PERCENT COARSER BY WEIGHT 3" 100 90 80 70 60 50 40 30 20 10 0 100 U.S. STANDARD SIEVE HYDROMETER 1-1/2" 1" 1/2" 1/4" #4 #10 #20 0 Mulch Gradation (Sieve Analysis) 10 20 30 40 50 60 70 80 90 10 1 GRAIN SIZE IN MILLIMETERS 100 0.1 Figure 17: Mulch size gradation by sieve analysis. 45 PERCENT FINER BY WEIGHT PERCENT COARSER BY WEIGHT 3" 100 90 80 70 60 50 40 30 20 10 0 100 U.S. STANDARD SIEVE HYDROMETER 1-1/2" 1" 1/2" 1/4" #4 #10 #20 0 18" Compost Gradation (Sieve Analysis) 10 12" Compost Gradation (Sieve Analysis) 20 30 40 50 60 70 80 90 10 1 GRAIN SIZE IN MILLIMETERS 100 0.1 Figure 18: Size gradation for two types of compost by sieve analysis The field density of compost and mulch material is variable along the installation, therefore; laboratory compaction testing was performed to determine the likely moist field density of the mulch and compost material. The compaction testing was performed by filling a deep pan with material using increasing amounts of force and agitation. The first compaction test was performed by loosely dropping the materials into the pan and leveling the top of the pan. The next two tests were performed by shaking the pan laterally while adding material in three and five layers. The final two tests were performed by shaking and compacting with hand pressure while adding material in three and five layers. Three test iterations were performed with the field compost and mulch samples, one check was performed on the cypress and mixed hardwood sample. The results shown in Figure 19 indicate that only small amounts of agitation 46 increase material density and that physical compaction does not significantly increase field density. Moist density (g/cc) 1 2 3 1 2 3 1 2 3 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 18" Compost 12" Compost Dropping only 3 layers, shake every layer 5 layers, shake every layer 3 layers, shake and hand pressure 5 layers, shake and hand pressure Field mulch Cypress Hard Figure 19: Determination of moist density. 3.1.2 Test soil Soil for the large-scale field testing of SCDs was red to reddish brown sandy clay stockpiled at the testing facility. Sieve and hydrometer gradations for two soil samples are shown in Figure 20. Sample 1 is representative of soil material used all tests performed except the 18" compost sock. Sample 2 was taken from the material used to test the 18" compost sock. This material was sourced from a different stockpile of similar red sandy clay soil. 47 PERCENT FINER BY WEIGHT PERCENT COARSER BY WEIGHT 3/8 100 90 80 70 60 50 40 30 20 10 0 10 U.S. STANDARD SIEVE HYDROMETER 4 10 20 40 60 100 200 0 Sample 1 (Sieve Analysis) 10 Sample 1 (Hydrometer Analysis) 20 Sample 2 (Sieve Analysis) Sample 2 (Hydrometer Analysis) 30 40 50 60 70 80 90 1 0.1 0.01 GRAIN SIZE IN MILLIMETERS 100 0.001 Figure 20: Test soil size gradation. 3.2 SCD field testing 3.2.1 ASTM D7351 test equipment Initial testing of the five selected sediment control devices was performed according to ASTM standard D7351, Standard Test Method for Determination of Sediment Retention Device Effectiveness in Sheet Flow Applications. This method assumes a 10-year, 6-hour storm event with a 100 mm (4 in) rainfall. Eroded soil is approximated for a 30 meter slope length using the Modified Universal Soil Loss Equation (MUSLE). Testing was performed using the ASTM D7351 test setup at the TRI Environmental Denver Downs Research Facility (DDRF) in Anderson, South Carolina. The test setup includes a large mixing tank, sloping (~21) fluid ramp, soil installation zone, downstream fluid ramp and downstream collection tank. Figure 21 shows the testing equipment used at DDRF with labels. 48 Figure 21: ASTM D7351 testing equipment at DDRF. The test area was prepared by compacting sandy clay soil in the soil installation zone, installing the SCD to be tested and mixing soil-water slurry. The test was performed by releasing the slurry at a controlled rate towards the SCD while making observations and collecting samples at regular intervals. Test soil was compacted in the soil installation zone using a jumping jack compactor and hand tamp. The surface of the soil zone was brought level to the edge of the upstream and downstream fluid ramps. The sediment control device to be tested was installed along the installation zone, generally centered, with an equal amount of soil exposed in front and 49 behind the SCD. ASTM D7351 specifies a range for test soil as shown in Table 11. The previous soil test gradation meets the test gradation. Table 11: Soil gradation indicated in ASTM D7351. Acceptable Range (mm) D100 < 25 0.5 < D85 <5.0 0.001 < D50 < 1.0 0.005 < D15 3.2.2 SCD installation Installation of each SCD was according to specifications in Georgia Soil and Water Conservation Commission (GSWCC) (2000). Installation procedures for each device are described in the following paragraphs. Geotextile silt fence was installed by placing the fabric in an 8-inch deep trench and then attaching the fence to wooden or steel posts. Each post was driven to a depth of 12 inches below the bottom of the trench using a large hammer. The low-flow (Type C) silt fence was stapled to a metal mesh backing that was attached to steel posts with two short segments of fence wire per post. The toe of the fence was configured into an L-shape by turning the bottom edge of the fabric in the upstream direction and covering with loose soil in accordance with the GSWCC guidance. The remaining portion of the trench was backfilled with the excavated trench soil and compacted on both sides of the silt fence. The approximate configuration of both geotextile silt fence installations is shown in Figure 22. Compost sock was delivered to DDRF pre-filled with compost and wrapped in a long coil. Before installation, the compost sock was unwrapped and a segment was cut to the length of the soil installation zone. The compost sock was positioned along the middle of the 50 installation zone and staked to the ground using 2-inch square stakes spaced 6 feet along the length of the installation. The arrangement of the stakes is shown in Figure 22. The sock was flattened with blows of a hand tamp and by walking along the top of the sock. Some soil was pressed into the upstream crevice between the bottom of the 12-inch sock and the soil surface. Loose compost was pressed into the upstream crevice of the 18-inch compost sock. Straw bales were installed in a 4-inch deep trench cut along the soil installation zone. The straw bales were positioned in the trench, end-to-end so that the twine-wrapped bale surfaces faced in the upstream and downstream directions. The bales were pressed together tightly and set in place by driving stakes through the bales and into the soil. Two stakes were driven through each bale, positioned in the center of the bale width, approximately 12 inches from the end of the bale. Remaining portions of the soil trench not filled by straw bale were filled with soil and compacted. Configuration of the straw bales is shown in Figure 22. Mulch berm was installed by setting small wooden stakes spaced approximately 4 feet apart, 6-inches from the back of the soil installation zone, as shown in Figure 22 . Pieces of long, approximately 10-inch wide planks of -inch plywood were propped against the stakes to form a short, reinforced vertical barrier. Loose mulch was distributing along the front of the plywood barrier and compacting using foot pressure. Configuration of the mulch berm is shown in Figure 22. 51 Soil Installation Zone Type A silt fence Type C silt fence Straw bales 12" Compost sock 18" Compost sock Mulch berm 52 Wooden stake Geotextile Metal stake Wire mesh Granular bentonite Loose compost Small wooden stake and planks Figure 22: Approximate configuration of installed SCDs. 3.2.3 SCD end connections The end connection of the SCD was an important detail for the project, as it was important to minimize the potential for stormwater to "by-pass" the SCD and flow (untreated) around the ends of the installation. Connection of the high and low silt fence to the test barrier panels at either end of the soil installation zone was achieved by wrapping the geotextile fabric along the barrier wall in the upstream direction. Wooden abutment stakes were driven in-line with the installed silt fence and immediately adjacent to the barrier wall. Between 6 and 12 inches of geotextile fabric was sandwiched between the barrier wall and the abutment stake. Silicone sealant was applied generously in the vertical void space between the geotextile and the wall panel. The abutment stakes were secured to the barrier wall with multiple wood screws. Direct connection between the barrier wall and the three-dimensional SCDs was not possible. For these installations, plywood wing walls were attached to the barrier panel with silicone sealant and screws and embedded in the soil installation zone upstream of the SCD installation. Each wall was approximately 18 inches long and embedded approximately 4 inches into the soil installation zone by trenching and placing and compacting soil. The wing walls were installed to form approximate interior 45 degree angles between the barrier wall and wing wall. To minimize flow of test water around the SCD installation, both wing walls were backfilled with granular bentonite clay during testing of the 12-inch compost sock and straw bales. During testing of the mulch berm and 18" compost sock, the wing walls were backfilled with mulch and loose compost, respectively. 53 3.2.4 ASTM D7351 test After installation of each SCD was complete, the test equipment was prepared for use. The test runoff water and soil mixture was prepared by filling the upstream mixing tank with 5,000 pounds of water from a ground water well on site. The test soil was sieved through a inch sieve and weighed using a portable scale then added to the mixing tank with the mixing blades in operation. Mix water was weighed using a four point truck scale under the upstream mixing tank. Approximately 5 minutes after adding the soil, a gate valve was partially opened to release the test fluid onto the upstream approach ramp. Release of the test fluid was monitored periodically by comparing the weight of the upstream tank with target weights based on an even release of 5300 pounds of combined soil and water over the 30 minute release period. The weight of test fluid passing the installed SCD was measured using the downstream collection tank and scale and recorded at 5 minute intervals. Samples of test water flowing through the upstream distributer (upstream samples) and flowing into the downstream collection tank (downstream samples) were collected in 500 mL, high-density polyethylene, sample containers at five minute intervals beginning five minutes after the first release from the upstream mixing tank. Each sample bottled was marked according to SCD being tested, sample time interval, and sample location. 3.3 Laboratory analysis 3.3.1 Turbidity Upstream and downstream samples were tested for turbidity using an Orbeco-Hellige TB200-10 portable turbidimeter with a measurement range of 0.01 to 1100 nephelometric turbidity units (NTU). Samples with turbidity greater than 1100 NTU were diluted with measured portions of deionized water until the sample was within the readable range of the 54 turbidity meter. A volume correction was applied to each diluted sample to approximate the actual turbidity reading. When calculating actual turbidity readings, turbidity of the de-ionized water was assumed to be zero; however, intermittent measurements of clear de-ionized water indicate the actual turbidity ranges between 0.1 and 5.0 NTU. 3.3.2 Total, suspended and dissolved solids Samples collected downstream of the installed SRD were filtered to determine total suspended solids (TSS) in accordance with EPA Method 160.2. Before filtering, each sample was thoroughly mixed by shaking. Filtering was performed by passing between 20 to 100 mL of sample water through clean glass fiber filter paper. The glass filter and retained soil material was oven dried and weighed. Total dissolved solids were determined by measuring electrical conductivity of the TSS filtrate. Conductivity readings were correlated to total dissolved solids (TDS) readings using a liner relationship. Upstream samples were tested for total solids (TS). Total solid testing was performed by oven drying the entire sample volume and recording sample weights before and after drying. 3.3.3 SCD nutrient retention The ability of the SCDs to capture nutrients was measured using cylinder filter testing. Samples of compost from the 12-inch and 18-inch socks, cypress and hardwood mulch and straw were placed in 12.7 cm (5 in) diameter 30.5 cm (12 in) long acrylic cylinders. The compost, mulch and straw samples were compacted to the field density indicated in previous material characterization tests. Initial nutrient content was determined by rinsing the materials with multiple 1 L portions of deionized water. Samples of the passing rinse water were collected after the first, fifth and ninth liter of water drained through each material. Nutrient retention capability was determined by filtering water with initial nutrient concentrations. Test water was prepared 55 by diluting nutrient stock solutions with deionized water to concentrations of 0.5 mg/L nitrite, nitrate, ammonia and phosphate. The test setup for nutrient retention is shown in Figure 23. Nutrient analysis included spectrophotometric tests for phosphorous, nitrite and nitrate and potentiometric measurement of ammonia. A list and description of each test method is included in Table 12. 56 Table 12: Nutrient test methods and description. Nutrient (Range) EPA Test Method Description Phosphorous, P (0.01 0.5 mg/L) 365.2 Antimony-phosph-molybdate complex reacts with to form blue color. Color intensity proportional to P content. Read at 650 nm. Nitrite-Nitrogen, NO2-N (0.01 1.0 mg/L) 354.1 Nitrite, sulfanilamide and Nethylenediamine dihydrochloride make red- purple complex. Read at 540 nm. Nitrate-Nitrogen, NO3-N (0.1 2 mg/L) 352.1 Nitrate reacts with brucine sulfate in acid at 100C to make yellow color. Absorbance read at 410 nm. AmmoniaNitrogen, NH3-N (0.03 1400 mg/L) 350.3 Potentiometric ammonia electrode measures ammonia diffusion through gas-permeable membrane. Fisher Scientific accumet ammonia combination electrode. 57 Dispersing container Clear test container Geotextile cover and bottom filter Tubing to hold geotextile Compacted test material Drain ports Collection container Figure 23: Laboratory filter test configuration. 3.3.4 Metal retention Metal testing was performed using the same filter device shown in Figure 23. Metal concentration of the initial 1 L portion of passing deionized water was measured. Metal retention capability was determined by passing and sampling 1 L of deionized water with 0.5 mg/L of lead, copper, and zinc concentrations and 1 L of rinse water through the filter materials. Initial 58 and retained metal concentrations were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES). ICP-OES testing was performed using axial readings. 3.4 Life cycle analysis (LCA) Life cycle analysis (LCA) of each sediment control device was performed using GaBi software version 6.0 by PE INTERNATIONAL. The GaBi software was used to generate LCA plans, balance life-cycle inventory and perform impact assessment according to USEPA TRACI methodology. 3.4.1 Goal and scope The purpose of the life cycle analysis is to compare the environmental impacts of five different SCD devices. Comparison of production, use and disposal phases of each SCD is also performed. 3.4.2 Functional unit (model conditions) The functional unit for the life cycle analysis is a 1,000 meter installation of SCD over a 1 year time period. Runoff and eroded soil is assumed to be generated by 11.5 10-year, 6-hour storm events. This was determined by dividing the mean yearly precipitation of 46 inches for Macon, Georgia (The Weather Channel 2012) into 4-inch storms. Nutrient and metal concentrations of runoff water is assumed as 0.5 mg/L. The model conditions equate to approximately 1.3 m3 per meter of installation per year. On the basis of erosion assumed in the ASTM D7351 test, the model conditions result in 348 kg of eroded soil generated per meter of installation per year. On the basis of on the assumed nutrient and metal concentrations, approximately 2.9 g of metal and nutrients are generated per meter of installation per year. 59 3.4.3 System boundaries System boundaries include production of all SCD materials, energy required for installation and maintenance, pollution from SCD effluent and end-of-life disposal. The system does not include energy and materials required for production of construction machinery. Figure 24 shows a general LCA flow chart and system boundaries. SYSTEM BOUNDARY PRODUCTION USE DISPOSAL MANUFACTURE EQUIPMENT INSTALLATION EQUIPMENT TRANSPORT EQUIPMENT RAW MATERIAL MATERIAL GENERATION SCD MANUFACTURE INSTALLATION MAINTENANCE (CLEANOUT) Figure 24: LCA system boundary 3.4.4 Processes A plan for each SCD was generated using a combination of processes developed by PE INTERNATIONAL distributed in their standard professional database and other processes developed using results of field and laboratory tests. Additionally, other processes were approximated based on fuel requirements and power performance from literature. Processes from within the PE INTERNATIONAL professional database are compilations of sub-processes that represent the majority of production and disposal steps included in the LCA. These processes and sub-processes are responsible for a significant amount of material and energy usage and are described in detail in the following paragraphs. The name of the process as shown on the plan pictographs is bolded and followed by the description. A use-phase process was developed for 60 each SCD using results of field and lab testing. Brief descriptions of the use phase processes are listed in Table 13; assumed conditions and generation of the use-phases is discussed in detail in Chapter 4. Unit processes for SCD installation and maintenance were approximated on the basis of simple energy requirements and are described in Table 14. European Union-27 Polypropylene fibers (PP) is an ISO compliant cradle-to-gate description of manufacture of polypropylene fibers compiled by PE INTERNATIONAL. Propylene from the cracking of refined crude oil is polymerized to polypropylene. The process assumes polymerization by gas-phase reactor methods. Fibers are formed by spinning; in this research spinning is assumed to be an acceptable model substitute to polypropylene strands by extrusion. Major inputs are water, crude oil, energy mix and various other elemental materials. The utilizable output is polypropylene fiber. Other process outputs as byproducts include radioactive waste, emissions to air and water from technosphere, waste heat, mixed hydrocarbons and heavy metals. A flow diagram of the processes included is shown in Figure 25. 61 Figure 25: Polypropylene fibers flow diagram (from GaBi 6.0). DE: Timber pine (40% water content) is an ISO compliant cradle-to-gate description of the generation of timber compiled by PE INTERNATIONAL. The utilizable output is timber with 40% moisture content. In this research, timber resulting from this process is assumed as wooden stakes used for staking SCDs. Required inputs are water, land use, energy mix, fuel for transportation sub-processes and various non-renewable elements. Output byproducts include radioactive and non-radioactive emissions to air and water, water from technosphere and hydrocarbons. Beneficial outputs from timber manufacture include oxygen production and erosion resistance. A descriptive flow diagram is included in Figure 26. 62 Figure 26: Timber process flow diagram (from GaBi 6.0) GLO: Steel wire rod is an ISO compliant cradle-to-gate process based on high quality worldsteel steel production data. This process describes the extraction of materials (i.e. coal, iron), processing (i.e. furnace) and forming required for steel production. Wire rod is considered a small-section strand coiled material. In this research, wire-mesh backing for Type C silt fence is considered as steel wire rod. Inputs for this process are air, water, coal, power, iron ore, steel scrap and various other renewable and non-renewable resources. The functional output is steel wire rod; output byproducts are sludge, slag, radioactive emissions, waste water and other hazardous and non-hazardous emissions to soil, water and air. GLO: Steel section is the same as the previous steel wire rod process with the utilizable output being hot-rolled steel sections instead of wire rod. Steel sections are considered I, H and 63 wide-flange beams and sheet-piling. In this research, steel fence posts for support of Type C silt fence are considered steel sections. GLO: Truck is an ISO compliant unit process compiled by PE INTERNATIONAL that describes the operation of a 3.3 metric ton (t) payload truck (7.5 t gross weight) for cargo transportation. In this research, transport of SCDs to the site and transportation of waste SCDs to landfill is assumed as a 3.3 t payload truck. Distances from SCD warehouse to site for each SCD plan is assumed as 10 kilometers (6.2 miles), distance from the site to landfill is assumed as 25 kilometers (15.5 miles). A 7.5 metric ton gross weight is roughly equivalent to a Ford F-550 truck. The corresponding FHWA gross vehicle weight rating (GVWR) is class 5 (medium duty) commercial. Inputs to the truck process are cargo at start and refined diesel fuel. Outputs are cargo at delivery and emissions to air from fuel combustion. The fuel requirement for this truck is roughly 0.004 kg diesel fuel per kg of payload per km travelled. US: Diesel mix at refinery is an ISO compliant cradle-to-gate process compiled by PE INTERNATIONAL that describes the production of refined diesel fuel used for transportation and electricity. Petrol refineries are complex plants that use different sub-processes to produce a combination of end-products of different quality. Sub-processes also differ based on the quality of raw petroleum. To represent diesel production in the United States, the utilization of separate refining processes is chosen to reflect the quality and quantity of product from 130 crude oil refineries located in the States. The refining procedure includes various distillation, hydrotreatment, conversion (i.e. cracking, coking) and finishing processes. The data set also includes elements of petroleum production such as crude oil exploration, well operation and transportation. Process inputs are water, crude oil and other renewable and non-renewable 64 resources. The utilizable output is diesel fuel; output byproducts include radioactive and nonradioactive waste, hydrocarbons, water from the technosphere and waste heat. EU-27: Landfill of untreated wood is an ISO compliant gate-to-grave process compiled by PE INTERNATIONAL that describes the cost of land-filling untreated wood. The process includes elements of municipal landfill construction (i.e. materials for clay and polyethylene barriers), operation (i.e., diesel for compactor, emissions from flare, partial reuse of methane), closure and maintenance. The process assumes a 30 m high landfill with cover area of 40,000 square meters that meets European code requirements for emissions. In this research, waste wooden stakes for installation of SCDs are assumed as waste wood. The utilizable input is untreated wood for landfill. Output byproducts are water and various emissions to air, water and soil. EU-27: Landfill of ferro-metals is a cradle-to-grave process similar to land-filling of wood waste with slightly different emissions that are associated with degradation of ferro-metal. EU-27: Landfill of plastic waste is a cradle-to-grave process similar to land-filling of wood waste with slightly different emissions that are associated with degradation of plastic. Land-filling sub-processes considered in both processes are shown in Figure 27. 65 Figure 27: Municipal landfill processes (GaBi 6.0) 3.4.5 Plans Processes relevant to the life cycle of each SCD are loaded into a GaBi plan editor. The processes are linked so that utilizable outputs from production stages (i.e., geotextile fabric, steel sections) are linked to corresponding use phases. When required, installation and maintenance processes are linked to use phases. Outputs from use phases become inputs for disposal phases. Transportation by truck is included as an intermediary between production, use and disposal phases. Emissions are byproduct outputs that are not transferred between life-cycle processes. Raw materials and emissions are compiled during the life-cycle balance. 66 Figure 28 through Figure 32 show life-cycle plans developed for each SCD analyzed. 67 Table 13: Inputs and outputs from use-phase processes Use-phase process Silt fence Type A Inputs (kg) Material required for installation of 1,000 meter length 109 polypropylene geotextile 426 timber (as stakes) Outputs (kg) Combined effluent passing 1,000 meter SCD installation over 1 year 4177 426 109 2.43 2.43 suspended solids wood for landfill plastic for landfill NO2, NO3, NH3, PO4 Cu, Pb, Zn Silt fence Type C 179 polypropylene geotextile 1737 steel sections (as stakes) 325 steel wire rod (as wire mesh) 6265 2062 179 2.14 2.14 suspended solids metal for landfill plastic for landfill NO2, NO3, NH3, PO4 Cu, Pb, Zn Compost sock 165 Polypropylene geotextile 65617 wood waste (as compost) 758 timber (as stakes) Straw bale 15626 straw 3031 timber (as stakes) Mulch berm 74803 wood waste (as mulch) 18535 suspended solids 758 wood for landfill 165 plastic for landfill 2.32 NO2 0.2 Cu 1.76 NO3 0.2 Pb 3.01 NH3 0.4 Zn 2.32 PO4 22972 suspended solids 3031 wood for landfill 0.63 NO2 1.1 Cu 0.20 NO3 1.1 Pb 3.51 NH3 1.6 Zn 19.5 PO4 13053 suspended solids 0.56 NO2 0.3 Cu 0.80 NO3 0.3 Pb 1.92 NH3 0.5 Zn 2.49 PO4 68 Table 14: Single-unit processes approximated for this research Process Description Geotextile Approximates weaving fibers into geotextile fabric. Assumes 2 kWh/kg manufacture power requirements for weaving (Palamutcu 2010). Inputs: Polypropylene fibers Output: Polypropylene geotextile (100% of input fibers by weight) Cleanout Mulch Trench Required effort to remove deposited soil mass from behind SCD devices. Assumes cleanout can be performed with 100kW (134 hp) excavator. This is a small hydraulic excavator with 0.7 to 0.8 m3 (22 28 ft3) bucket capacity. Roughly equivalent to a CAT 318E. Inputs: 0.172 kg diesel per 1,000 kg excavated material Outputs: Combustion emissions Approximates mulching site coverage. Assumes mulcher is 63 kW (85 hp) tractor or skid-steer with mulch cutting attachment. Roughly equivalent to a Bobcat S-750. Inputs: 0.108 kg diesel per 200 kg mulch generated Outputs: Wood waste (mulch), combustion emissions Required effort for trenching and installation of silt fence material. Modeled as a 46kW (61 hp) small tractor or skid-steer such as a Bobcat S-160. Inputs: 0.078 kg diesel per 1000 kg excavated material Outputs: Excavated material, combustion emissions Straw bale creation Approximates the combined energy required for cultivation and baling straw bales. Assumes fuel usage for cultivation and baling is 0.60 and 0.45 US gallons, respectively (Downs and Hansen 1998). For a density of 6.1 lb/gallon and assuming 25 bales produced per acre, combined fuel consumption is 0.193 kg per m of bale installation. Inputs: Straw, diesel fuel as described Outputs: Straw bale 69 Figure 28: Process plan for silt fence Type A 70 Figure 29: Process plan for silt fence Type C 71 Figure 30: Process plan for compost sock 72 Figure 31: Process plan for straw bale barriers 73 Figure 32: Process plan for mulch berm 3.4.6 Impact analysis LCA impact analysis was performed using GaBi 6.0 inventory balance functions and grouping and impact categories developed by the EPA Tool for the Reduction and Assessment of Chemical and other environmental Impacts (TRACI). TRACI is previously discussed in Chapter 2. All inventory balances and impact sorting was performed using GaBi 6.0. 74 RESULTS AND ANALYSIS 4.1 SCD field performance Seven (7) total large-scale SCD installations were tested using the ASTM D7351 test method during this study. The tests are briefly described in Table 15. The initial test of Type C silt fence (test #2) resulted in a significant undercutting failure at the west installation barrier connection. The fence installation was repaired and retested (test #2b). Testing of 12" compost sock (test #3) resulted in significant undercutting along approximately of the length of the sock installation. Compost sock was retested using a larger, 18" diameter sock (test #6). Table 15: Summary of ASTM D7351 testing. Test Number (test date) 1 (8/6/12) Material Tested Silt fence (Type A) Comments Successful test 2 (9/12/12) High-flow silt fence (Type C) SCD failed at 18 minutes from severe undercutting (blow-out) 2b (9/14/12) Low-flow silt fence (Type C) Successful retest of low-flow fence 3 (9/21/12) 12" Compost sock Poor performance, significant undercutting of SCD installation 4 (9/24/12) Straw bales Significant flow between bales 5 (9/26/12) Mulch berm Successful test 6 (2/23/13) 18" Compost sock Successful test with some overtopping 75 Type A silt fence during and after testing is shown in Figure 33, respectively. Figure 33 shows the significant amount of sediment that settles before reaching the fence installation. At termination, flow through the fence had decreased significantly. Connection of the geotextile fabric to the test barrier wall is by wrapping the fabric upstream and pinching between abutment post and wall with added sealant at post-to-fabric and fabric-to-wall interfaces. Figure 33: Type A silt fence during (left) and after (right) test. 76 Low-flow silt fence testing during and the accumulated sediment behind the fence after testing is shown in Figure 34. Figure 34 shows the retest of the Type C fence. Figure 34: Type C silt fence during (left) and after (right) test. 77 Twelve (12)-inch compost sock is shown during and after testing in Figure 35. Streaking behind the sock installation is where significant undercutting occurred during testing. No connection was possible between the compost sock and the barrier walls. Flow around the sock at the installation ends was reduced by installing wing-walls with granular bentonite backfill between the sock and wing-wall. Figure 35: 12-inch compost sock during (left) and after (right) test. 78 The installed line of straw bales is shown before and after testing in Figure 36. Soil present on the upstream and downstream ramps was removed prior to the start of the test. Flow around the bales was limited using wing-wall and granular bentonite barriers. Figure 36: Straw bales before (left) and after (right) test. 79 The mulch berm is shown during and after testing in Figure 37. The mulch berm was supported using plywood and small wooden posts. Flow around the berm was limited using wing-walls backfilled with compacted mulch. Figure 37: Mulch berm before (left) and after (right) test. 80 The 18-inch compost sock is shown during and after 7351 testing in Figure 38. Flow around the sock was limited using wing-walls backfilled with compacted compost taken from extra sock material. Overtopping of the compost sock occurred at 15 minutes after the start of the test. Figure 38: 18-inch compost sock before (left) and after (right) test. 81 4.1.1 Soil and water retention Weights of the upstream mix tank and downstream collection tanks were recorded at intervals during the ASTM D 7351 testing. The weight of the soil and water mixture retained at the SCD installation is given as the total test weight less the weight of both tanks expressed as: = 5300 - - Where WSCD is the weight retained at the SCD, Wup and Wdown are the recorded mix and collection tank weights, respectively, and 5,300 lbs is the total soil-water mixture test weight. Recorded tank weights and SCD retention for the seven tests conducted are shown in Figure 39 through Figure 45. Weight (lbs) 6000 5000 4000 3000 2000 1000 0 0 Mix Tank Collection tank Retained at Type A Fence 20 40 60 80 100 Elapsed Time (min) Figure 39: Recorded tank weights and retention of Type A silt fence. 82 Weight (lbs) 6000 5000 4000 3000 2000 1000 0 0 Mix Tank Collection Tank Retained at Type C Silt Fence 20 40 60 80 100 Elapsed Time (min) Figure 40: Recorded tank weights and retention of high-flow Type C silt fence with failure from undercutting at 18 minutes. 6000 5000 Mix Tank Collection Tank Retained at Type C Fence 4000 Weight (lbs) 3000 2000 1000 0 0 20 40 60 80 100 Elapsed Time (min) Figure 41: Recorded tank weights and retention of high-flow Type C silt fence. 83 Weight (lbs) 6000 5000 4000 3000 2000 1000 0 0 Mix Tank Collection Tank Retained at 12" Compost Sock 20 40 60 80 100 Elapsed Time (min) Figure 42: Recorded tank weights and retention of 12-inch compost sock with significant undercutting of beginning at 5 minutes. 6000 5000 4000 Mix Tank Collection Tank Retained at 18" Compost Sock Weight (lbs) 3000 2000 1000 0 0 20 40 60 80 100 Elapsed Time (min) Figure 43: Recorded tank weights and retention of 18-inch compost sock. 84 Weight (lbs) 6000 5000 4000 3000 2000 1000 0 0 Mix Tank Collection Tank Retained at Straw Bale 20 40 60 80 100 Elapsed Time (min) Figure 44: Recorded tank weights and retention of straw bales. 6000 5000 4000 Mix Tank Collection Tank Retained at Mulch Berm 3000 2000 1000 0 0 20 40 60 80 100 Elapsed Time (min) Figure 45: Recorded tank weights and retention of mulch berm. Weight (lbs) 85 The flow-through rate or exit flow rate from the SCD installations is approximated as the change of collection tank weight over time intervals between record times expressed as: = 2 2 - - 1 1 Where W/t is the rate of weight increase in the collection tank and W2 and W1 are tank weights recorded at times t1 and t2, respectively. Maximum exit rates for each test are shown in Figure 46. The highest rates were calculated for the initial test of Type C silt fence and the 12- inch compost sock, both tests failed by undercutting. Successful tests of both silt fence, much berm and 18-inch compost sock experienced maximum exit rates between 177 and 204 lb/min. 86 1 2 2b Maximum SCD Exit Rate (lb/min) 0 50 100 150 200 250 300 350 400 450 Type A silt fence 203 Type C silt fence* 297 Type C silt fence 177 12" compost sock** 380 Straw bales 242 Mulch berm 188 18" compost sock 204 3 4 5 6 *Fence failure by undercutting/blowout **Significant undercutting occured below Figure 46: Maximum exit rates for each SCD tested. 4.1.2 TSS reduction Results summarized in section 4.1.1 show the ability of the SCD installations to retain the test soil and water mixture and do not indicate the amount of solids captured by the devices. To determine the removal capability of each device to remove suspended solids, TSS concentration of water entering the collection tank was determined. Figure 47 through Figure 53 show the TSS results for five minute samples collected during each test. Two TSS tests were performed for each sample. The following bullets discuss some of the trends and irregularities visible in/between the figures. 87 1. TSS values generally increase during the 30 minute discharge period as soil-laden test water accumulates behind the installations. 2. TSS values decrease substantially after the initial discharge period. 3. Silt fence installations (Figure 47, Figure 48, and Figure 49) show a slight decrease in TSS values during the discharge period. The decrease is due to formation of a filter cake of soil particles at the upstream face of silt fence. 4. The TSS peak at 20 minutes for the 18" compost sock (Figure 51) corresponds to overtopping of the installation. 5. Decreasing TSS in the after 20 minutes for the 18" compost sock (Figure 51) and for the mulch berm (Figure 53) is due to ripening of the filter material. 6. Increased TSS at 15 minutes for the 12" compost sock (Figure 50) and straw bale (Figure 52) correspond to undercutting of the sock and penetration at the bale interfaces. The following figures show the results for two iterations of TSS filter tests performed with the same samples. The high repeatability of separate TSS tests is not to be confused with iterations of field testing procedures, which would produce charts with higher variability. 88 Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 Type A silt fence Type A silt fence copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 47: Measured TSS downstream of Type A silt fence, 5-minute samples. Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 Type C silt fence with failure Type C silt fence with failure copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 48: Measured TSS downstream of Type C silt fence, 5-minute samples. Fence failed 18 minutes into test. 89 Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 Type C silt fence Type C silt fence copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 49: Measured TSS downstream of Type C silt fence retest, 5-minute samples. Total suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 12-inch compost sock 12" compost sock copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 50: Measured TSS downstream of 12-inch compost sock, 5-minute samples. 90 Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 18" compost sock 18" compost sock copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 51: Measured TSS downstream of 18-inch compost sock, 5-minute samples. Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 Straw bales Straw bales copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 52: Measured TSS downstream of straw bales, 5-minute samples. 91 Total Suspended solids, TSS (mg/L) 16000 14000 12000 10000 8000 6000 4000 2000 0 0 Mulch berm Mulch berm copy 1 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 53: Measured TSS downstream of mulch berm, 5-minute samples. The minimum, maximum and average value of the averaged TSS test pairs are shown in Table 16. The Type A silt fence installation showed the lowest maximum TSS value. Measured TSS values for each sample collected are included in Appendix B. Table 16: Range and average TSS for each test. Measured TSS (mg/L) Test Device Minimum Maximum Average 1 Type A 33 2531 532 2 Type C 80 15078 6261 2b Type C 25 4911 807 3 12" Sock 2411 12447 7418 4 Straw Bales 316 8675 4195 5 Mulch Berm 354 4857 2856 6 18" Sock 750 8749 4324 92 4.1.3 SCD removal efficiency The amount of solids passing each SCD device is a function of the flow rate through the SCD device and the downstream solids concentration. The amount of solids passing the SCD between times t1 and t2 is approximated as the volume passing the SCD over the time range multiplied by the average TSS concentration for the same time range and is expressed as: = = 2 - 1 2 - 2 1 Where Ws is the weight of solids passing the SCD, V is volume passing the SCD approximated as the increase of tank weight (W2 W1) divided by the unit weight of water (w). W2 and W1 are the collection tank weights and C2 and C1 are the measured TSS concentrations at times t1 and t2, respectively. The assumption that the passing volume is equal to the change in weight over the unit weight of water is validated by showing the negligible increase in unit weight by the maximum measured TSS concentration (15,078 mg/L): = (1 - ) + = = = 1 (1 - 15,078 2.65 111016) + 2.65 1 15,078 ( 2.65 111016) = 1.0094 1 Where * is the corrected unit weight, w and s are the unit weights of water and solid (assumed as 2.65w), Vs and Ws are the solid volume and weight, respectively, and Cs is the maximum measured TSS concentration for downstream samples. Figure 54 shows cumulative solids passing each SCD tested. Incremental TSS passing each SCD is minimal at the end of the test 93 duration except for the failed 12-inch compost sock installation. Additional testing time for the 12-inch sock would yield a significantly higher cumulative downstream TSS value. TSS passing SCD (lbs) 40 35 30 25 20 15 10 5 0 0 10 20 30 40 50 60 Elapsed Time (min) Type A silt fence Type C silt fence Straw bales 18" compost sock Type C silt fence (failed) 12" compost sock Mulch berm Figure 54: Cumulative solids passing SCDs. Sediment removal efficiency of each SCD can be approximated as the ratio of solids retained to total solids tested. Removal efficiency for this test is expressed as: (%) = 100 - 100 Where EFs is the solids removal efficiency, Wp is the cumulative weight of soil passing the SCD and Ws is the total weight of solids tested, 136.1 kg (300 lbs). Calculated removal efficiencies for 94 each SCD tested are shown in Figure 55. Additional test time for the 12-inch sock would have resulted in higher downstream TSS; therefore, the removal efficiency shown is not accurate. TSS Removal Efficiency (%) 100 98.4 98 97.6 96 95.0 94 92.9 92 91.6 91.2 90 88.2 88 86 84 82 Type A silt Type C silt fence fence Mulch berm 18" Failed type Compost C silt fence sock Straw bales 12" Compost sock Figure 55: Removal efficiency including solids removed by sedimentation. 4.1.4 Turbidity reduction Measured turbidity of samples made downstream of the Type A and C silt fence installations is shown in Figure 56. Turbidity of samples taken downstream of the 12 and 18inch compost sock installations is shown in Figure 57. Turbidity of samples taken downstream of the straw bale and mulch berm installations is shown in Figure 58. Each test shows a decrease in turbidity after the initial 30 minute release period. The failed Type C silt fence installation resulted in the highest downstream turbidity measurements. 95 12000 10000 8000 6000 4000 Type A silt fence Type A silt fence copy 1 Type C silt fence Type C silt fence copy 1 Type C silt fence with failure Type C silt fence with failure copy 1 Downstream Turbidity (NTU) 2000 0 0 20 40 60 80 Elapsed Time (min) Figure 56: Measured turbidity downstream of Type A and Type C silt fence installations. 96 Downstream Turbidity (NTU) 12000 10000 8000 6000 12" Compost Sock 12" Compost sock copy 1 18" Compost Sock 18" Compost sock copy 1 4000 2000 0 0 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 57: Measured turbidity downstream of 12 and 18 inch compost socks. 97 12000 10000 8000 6000 Straw bale Straw bale copy 1 Mulch berm Mulch berm copy 1 Downstream Turbidity (NTU) 4000 2000 0 0 10 20 30 40 50 60 70 80 90 Elapsed Time (min) Figure 58: Measured turbidity downstream of mulch berm and straw bales. Figure 59 shows turbidity measurements of upstream samples collected at the distributor output from the mixing tank. Turbidity of upstream samples generally ranges between 7,500 to 12,500 NTU except for samples taken during the 18" compost sock test. High turbidity of the 18" compost sock test may be a result of greater fines content of the new stockpile soil. Average upstream turbidity was calculated for each test and is shown in Table 17. Measured turbidity values for each sample collected are included in Appendix C. 98 Upstream Turbidity (NTU) 25000 20000 15000 10000 5000 0 0 5 10 Type A silt fence Straw bale 15 20 Elapsed Time (min) Type C silt fence Mulch berm 25 30 35 12" Compost Sock 18" Compost Sock Figure 59: Measure turbidity of upstream samples. Table 17: Average upstream turbidity measurements. Test Number Material Average Upstream Turbidity (NTU) 1 High-flow silt fence 2 Low-flow silt fence with failure 2b Low-flow silt fence 6,386 Not tested 9,298 3 12" compost sock 11,077 4 Straw bales 10,932 5 Mulch berm 10,990 6 18" compost sock 17,482 Turbidity reduction as a percentage of upstream average turbidity is show in Figure 60. The figure shows three reductions, the average reduction for all test samples, the average reduction during the initial 30 minute release period and the reduction for the remaining time after the initial release period. 99 Turbidity Reduction (%) 100 92.3 92.8 90 80 73.8 70 61.9 64.9 60 49.2 50 40 30 20 Type A silt Type C silt 12" compost Straw bales Mulch berm 18" compost fence fence sock* sock 1 2b 3 4 5 6 Test duration First 30 minutes Time after 30 minutes *Test terminated after 30 minutes Figure 60: Average turbidity reduction for entire test, first 30 minutes and time after first 30 minutes. A good agreement between measured turbidity and total suspended solids (TSS) was apparent for tests of silt fence, mulch berm and 18" compost sock and is shown in Figure 61. Measurements taken from samples downstream of the failed Type C silt fence show much higher TSS and turbidity readings than samples downstream of the 18" compost sock, mulch berm, and Type A silt fence. TSS versus turbidity for tests of straw bale and 12" compost sock are shown in Figure 62 with the trend from Figure 61. Samples from the 12" sock with low turbidity and high TSS values were taken between 15 and 30 minutes, after failure of the sock by undercutting. The failure resulted in coarse material passing the installation; coarse material increases TSS but has little affect on turbidity measurements. Four samples downstream of the straw bale 100 installation showed large turbidity values with small TSS. These samples were taken between 35 and 50 minutes. Only a small amount of water passed the straw bale installation after the initial 30 minute discharge period. These samples consist of heavy slurry of fine particles still in suspension, slowly leaking through the straw bale connections. Downstream TSS (mg/L) 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 0 y = 1.3259x - 321.37 R = 0.9506 Type A silt fence Type C silt fence with failure Type C silt fence Mulch berm 18" Compost sock Linear (5 tests) 2000 4000 6000 8000 10000 12000 14000 Downstream Turbidity (NTU) Figure 61: Correlation for turbidity and TSS. 101 Downstream TSS (mg/L) 18000 16000 14000 12000 10000 8000 6000 4000 2000 0 0 12" compost sock Straw bale Expected Trend 5000 10000 Downstream Turbidity (NTU) Figure 62: Turbidity vs. TSS for straw bale and 12" compost sock. 4.1.5 Total dissolved solids (TDS) reduction Calculated TDS values for downstream samples for each SCD tested are shown in Figure 63. TDS values are somewhat erratic; however, an initial decrease in TDS seems to occur for the three-dimensional devices. TDS values for Type A silt fence are fairly uniform over the duration of the test. TDS values for the Type C silt fence generally increase over the duration of the test. No correlation between TDS and TSS or turbidity was apparent. 102 Total Dissolved Solids, TDS (mg/L) 180 170 160 150 140 130 120 110 100 0 Type A silt fence Type C silt fence with failure Type C silt fence 12" Compost sock Straw bale Mulch berm 20 40 60 Elapsed Time (min) 80 100 Figure 63: Measured TDS for each SCD tested. 4.1.6 SCD installation Good performance of sediment control barriers depends on proper installation. Installation procedures of the devices tested were significantly different. The following bullet points are observations of SCD installation made during field testing. 1. Installation of the mulch berm was easy and fast. 2. Mulch was easier to place than compost socks. 3. Compost sock installation was not difficult once the socks were in the soil installation zone; however, moving the socks into the zone was labor intensive, even with help from a forklift and operator. 4. On site filling of compost sock with a compost blower would make installation easier. 5. Straw bale installation was difficult due to the required 4-inch deep trench that had to be excavated. 103 6. Installation of silt fence required significant trenching. Type C silt fence was slightly more difficult than Type A silt fence due to the metal mesh backing. No machinery was used during this research; Table 18 lists machinery commonly used for installation of SCDs. Table 18: Available SCD installation equipment SCD Material Silt fence Compost sock Straw bales Mulch berm Available Installation Equipment Line trencher Compost blower, fork lift Mulcher, chipper, grading equipment 4.1.7 Discussion Results from the ASTM 7351 field testing are generally good; however, the test method has multiple shortcomings with regards to this study. The most significant failing is the large soil loading used during testing. The sediment load described in the standard (136 kg soil in 600 gallons water) is approximately 60,000 mg/L. After correcting for soil water content, the actual sediment load tested in this study was approximately 45,000 mg/L. The sediment load is the load from erosion during a 10-year, 6-hour storm event on a 30 meter long slope using the MUSLE. According to the standard, the same size storm is used for sizing detention ponds. The storm choice is adequate for worst-case testing, but is not a good choice for typical conditions over a year. The assumed storm for the LCA study is discussed in detail in section 4.3.4 Discussion. Another limitation of the test method is that variable soil gradations will behave differently during the test. Coarse grains will settle faster and fine particles will stay in suspension longer. 104 Also, fine particles will form a filter cake at the upstream filter. Test method ASTM D7351 includes a gradation for test soil, but there is a large range of acceptable grain sizes. Additionally, the soil selected for the study may not truly represent the soil fraction that would erode during a storm event. Sloping soil bed tests with simulated rainfall performed by Faucette et al. (2009) includes only the eroded fraction of soil and compares results to control plots with no SCDs. These tests require multiple iterations to demonstrate consistent performance of erosion occurring in control plots. Additional weaknesses of the test method are that edge effects are difficult to overcome using the method as specified and SCD installation will vary between studies. A curved installation zone similar to McFalls et al. (2009) will eliminate edge effects. Adequate installation is difficult to control. TSS removal efficiency for silt fence measured during this study generally agrees with values determined by Beighley and Valdes (2009) and slightly higher than Barrett et al. (1998). Removal efficiency for compost socks is slightly higher than measured by Faucette et al. (2009). Removal efficiency for straw bales and mulch berm is significantly higher than Faucette et al. (2009). An improvement to the solids retention testing in this study is to perform duplicate field tests for a variety of water and soil loadings (as calculated for different storm events). Test iterations on this scale will take a significant amount of effort. This study shows that silt fence performed the best with respect to reducing downstream turbidity and TDS. Although the performance of SCDs with regards to turbidity and TDS were measured, no good method to include the results in the LCA was determined. 105 4.2 SCD laboratory performance Filter testing of compost, mulch and straw material was performed by passing multiple 1 liter (L) increments of deionized water (DI) or prepared nutrient and metal solutions through the laboratory filter containers. The schedule of filter testing is indicated in Table 19. Table 19: Schedule of laboratory testing Test Increment Description 1 1 L DI Initial rinse, sample collected 2 1 L DI 3 1 L DI 4 1 L DI 5 1 L DI Sample collected 6 1 L DI 7 1 L DI 8 1 L DI 9 1 L DI Sample collected 10 1 L 0.5 ppm NO2, NO3, NH3, PO4 - Sample collected 11 1 L DI Sample collected 12 1 L 0.5 ppm Pb, Zn, Cu Sample collected 13 1 L DI Sample collected Nutrient levels of each sample were tested with an initial aliquot sized as needed to decrease turbidity and lower suspected high nutrient concentrations to within testing limits. One repetition was performed for each test. Variance between initial and replicate tests for concentrations up to 2 mg/L is shown in Figure 64. Tests for nitrite and orthophosphate showed good agreement between approximately 0.25 mg/L to 2 mg/L. Nitrite results are more variable at lower concentrations, shown in Figure 65. Potentiometric ammonia measurement shows high variability at all concentrations. 106 Concentration 2 (mg/L) 2.0 1.5 1.0 0.5 0.0 0.0 0.5 1.0 1.5 2.0 Nitrite Nitrate Ammonia Orthophosphate 1:1 Concentration 1 (mg/L) Figure 64: Initial and replicate nutrient results, 0 2 mg/L Concentration 2 (mg/L) 0.5 0.4 0.3 0.2 0.1 0.0 0.0 0.1 0.2 0.3 0.4 0.5 Concentration 1 (mg/L) Nitrite Nitrate Ammonia Orthophosphate 1:1 Figure 65: Initial and replicate nutrient results, 0 0.5 mg/L Initial nutrient concentrations measured in the first portion of water passing the filters are shown in Figure 66. Leachate collected from the compost samples generally shows the highest initial nutrient concentrations. The mixed hardwood mulch and straw samples contained very high levels of phosphate. Nutrient concentrations of leachate water after rinsing with a total of 9 L of deionized water are shown in Figure 67. The source of increased nutrient values measured for the 18" compost sample is not clear. For simplicity, data used for characterizing nutrient 107 retention is limited to results from tests with the 12" compost sample. Also, results for both mulch samples are combined to describe a mixed mulch material. Measured nutrient concentrations for each test and material are included in Appendix D. 4.2.1 Nutrient retention Nutrient retention is determined by filtering water with an initial 0.5 mg/L concentration through each filter container. Concentrations of test water were checked prior to testing; generally, the test water was within 0.025 mg/L of the target 0.5 mg/L concentration. Nutrient retention is calculated as the measured initial nutrient concentration less the nutrients passing the filter container. Results of nutrient retention tests are shown in Figure 68. A negative value indicates nutrients were added to the test water after passing through the filter container. A positive value indicates nutrients were removed from the test water. 5 25.5 32.9 +/- 1.1 4 Concentration (mg/L) 3 NO2 NO3 NH3 2 TN PO4 1 0 18" Compost 12" Compost Cypress Mulch Hard Mix Mulch Straw Figure 66: Initial nutrient concentrations (mg/L) 108 Concentration (mg/L) 5 6.5 9.8 4 3 2 1 NO2 NO3 NH3 TN PO4 0 18" Compost 12" Compost Cypress Mulch Hard Mix Mulch Straw Figure 67: Nutrient concentrations after ninth liter of water 109 PO4 TN -0.097 NH3 NO3 Compost NO2 0.041 0.096 0.040 0.152 PO4 0.000 TN NH3 0.115 NO3 NO2 0.339 0.387 0.841 Mulch -4.058 Straw PO4 TN -0.320 NH3 NO3 NO2 0.487 0.453 0.354 Figure 68: Nutrient retention of compost (upper), mulch (middle) and straw (lower). Negative is nutrient (mg/L) added to stream, positive is nutrient (mg/L) removed from stream. 110 4.2.2 Metal retention Metal test water was prepared by mixing deionized water with concentrated metal standards to form approximate 0.5 mg/L concentrations of lead (Pb), copper (Cu) and zinc (Zn). Tests of the initial standard mix indicate metal concentrations of 0.465 mg/L 0.01 mg/L. Metal retention is the initial test concentration less the concentration measured in water passing the filter containers. Metal concentrations retained in the containers are shown in Figure 69. Each material retained a significant portion of the test metal. Measured metal concentrations for each test and material are included in Appendix E. 0.5 Metal Concentration (mg/L) 0.4 0.3 Cu 0.2 Zn Pb 0.1 0 18" 12" Cypress Hard Mix Compost Compost Mulch Mulch Straw Figure 69: Metals retained in filter material 4.2.3 Discussion Multiple iterations of nutrient testing were performed in an attempt to eliminate interference from sample turbidity. When included in the life cycle analysis; however, the most important result of the nutrient retention testing is that straw adds a significant amount of phosphorous to the water flow. An improvement to nutrient and metal retention testing would be 111 to include soil particles in the test water and measure nutrient/ metal retention for geotextile silt fence, assumed to be zero in this study. 4.3 Life cycle analysis 4.3.1 Use-phase description The purpose of the field and laboratory testing was to generate descriptive use phase processes for each SCD. Overall soil, nutrient and metal retention characteristics of the silt fence, compost, mulch and straw tested is summarized in Table 20. Negative values indicate nutrients were added to the water flowing through the material. The retention of soil solids is based on performance during the standard ASTM D7351 test method. Test conditions are as described in Chapter 3. Table 20: Percentage soil, nutrient or metal retained Material Soil NO2- NO3- N N NH3 PO4-P Cu Zn Pb Type A 98.4 Type C 97.6 Compost 92.9 8.0 30.4 -19.3 -63.9 93.8 85.9 91.7 Mulch 95.0 77.4 67.8 23.0 -0.1 89.0 79.6 87.2 Straw 91.2 70.8 90.6 -63.9 -811.6 47.0 26.0 47.6 Note: Blank cells not tested, assumed as zero Inputs required for each use-phase process are calculated by multiplying the materials required per unit length listed in Table 9 by the 1,000 meter test length. Outputs over the life of the SCD installation are determined by multiplying the retention values in Table 20 by the 112 expected total runoff approaching the installation for the 1-year test duration. The amount of solids passing each SCD is calculated as: = 136 11.5 1,000 4.5 (%) The predetermined solid mass generated per storm event is 136 kg, 11.5 storms per year is the mean precipitation for Macon, Georgia (46") divided by the storm precipitation (4"), 4.5 m is the length of the test installation and R is the solids retention of each SCD. The amount of nutrients and metals passing each SCD is calculated as: 1,000 / = 0.5 11.5 4.5 (%) The concentration of nutrients and metals in the runoff stream is assumed as 0.5 mg/L, Q is the effluent runoff water passing each SCD and R is the nutrient and metal retention. The resulting outputs for the LCA model installation is listed for each SCD and contaminant in Table 21. Table 21: Solids, nutrients and metals (kg) leaving site over 1 year period. Solids NO2-N NO3-N NH3 PO4-P Cu Zn Pb Silt fence Type A Silt fence Type C Compost Mulch 5,575.8 2.4 2.4 2.4 2.4 2.4 2.4 2.4 8,363.6 2.1 2.1 2.1 2.1 2.1 2.1 2.1 24,742.4 2.3 1.8 3.0 2.3 0.2 0.4 0.2 17,424.2 0.6 0.8 1.9 2.5 0.3 0.5 0.3 Straw 30,666.7 0.6 0.2 3.5 19.5 1.1 1.6 1.1 113 4.3.2 Inventory balance The following charts show inventory balances performed using GaBi 6.0 life cycle analysis software. The balances are performed using the life-cycle maps shown in Chapter 3. The software scales required production units (i.e. plastic, metal, and timber ) and disposal units (i.e. wood, plastic, and steel waste) to meet inputs required and outputs generated by the use phases developed previously. The contribution from each SCD to runoff solids is shown in Figure 70. Figure 71 shows energy requirements for each SCD. Figure 72 shows energy requirements for only the production processes. The large amount of timber required for staking straw bales is energy intensive with a majority of the energy being renewable from solar. As expected, production of steel sections is energy intensive and mostly from non-renewable sources. Polypropylene is shown as somewhat energy intensive from mostly non-renewable sources. Figure 73shows non-renewable resources used during production phases. As expected, steel production requires a large amount of nonrenewables, generally followed by polypropylene and then timber. 114 Mass [kg] GaBi diagram:soil loss from 'Construction industry' gfedcb Soil loss by erosion into water 30,000 25,000 20,000 15,000 10,000 5,000 0 Use Phase Straw Bal... Use Phase Mulch Ber... Use Phase Silt fence ty... Use Phase Compost S... Use Phase Silt Fence t... Figure 70: Soil loss balance Energy (net calorific value) [MJ] GaBi diagram:aggregated - Inputs Non renewable energy resources Renewable energy resources 65,000 60,000 55,000 50,000 45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 Silt fence, type C Compost sock Straw Bale Barrier Silt fence, type A Mulch Berm Figure 71: Energy resources by SCD 115 Energy (net calorific value) [MJ] GaBi diagram: energy diagram for 'production phases' 50,000 gfedcb Non renewable energy resources gfedcb Renewable energy resources 45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 straw, timber type C, PP compost, timber... type C, steel wire type C, steel sections compost ,PP type A, PP type A, timber Figure 72: Energy by production phase GaBi diagram: non-renewable resources in ''Production' phase 9,000 gfedcb Non renewable resources 8,000 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0 GLO: Steel secti... DE: Timber pine ... EU-27: Polyprop... EU-27: Polyprop... GLO: Steel wire ... EU-27: Polyprop... DE: Timber pine ... DE: Timber pine ... Figure 73: Non-renewable resources Mass [kg] 116 Emissions to freshwater from each SCD not including runoff and waste water from production and manufacturing processes or eroded soil are shown in Figure 74. Figure 75 shows the top six inorganic emissions to fresh water. The large values of phosphorous, nitrate, nitrite and ammonia are due to the nutrient levels in runoff water. Figure 76 shows emissions to air associated with each SCD. GaBi diagram: other emissions to fresh water gfedcb Inorganic emissions to fresh water gfedcb Heavy metals to fresh water gfedcb Solids (suspended) gfedcb Analytical measures to fresh water gfedcb Organic emissions to fresh water gfedcb Metals (unspecified) gfedcb Silicon dioxide (silica) 45 40 35 30 25 20 15 10 5 0 Straw Bale Barrier Compost sock Silt fence, type C Silt fence, type A Mulch Berm Mass [kg] Figure 74: Freshwater emissions. Soil and water (runoff and process) not included 117 Mass [kg] GaBi diagram: in-organic emissions to water gfedcb Chloride gfedcb Phosphate gfedcb Ammonia gfedcb Nitrite gfedcb Sodium (+I) gfedcb Nitrate 20 18 16 14 12 10 8 6 4 2 0 Straw Bale Barrier Compost sock Silt fence, type C Silt fence, type A Mulch Berm Figure 75: Top 6 in-organic emissions to fresh water GaBi diagram: emissions to air g f e d c b Water (evapotranspiration) g f e d c b Water vapour g f e d c b Carbon dioxide g f e d c b Oxygen g f e d c b Carbon dioxide (biotic) 30,000 25,000 20,000 15,000 10,000 5,000 0 Straw Bale Barrier Silt fence, type C Compost sock Silt fence, type A Mulch Berm Figure 76: Top 5 emissions to air Mass [kg] 118 4.3.3 Life cycle impact analysis (TRACI) Results of the impact analysis performed with GaBi 6.0 according to US EPA's TRACI methodology are shown for each SCD in the following figures. Figure 77 shows the global warming potential (GWP) of each SCD in terms of equivalent kg of CO2. Figure 78 shows acidification potential (AP) of each SCD in terms of equivalent kg of H+ ions. Figure 79 shows eutrophication potential (EP) of each SCD in terms of equivalent kg of nitrogen (N). Figure 80 shows ecotoxicity to water of each SCD in terms of m3-days per kg potential affected fraction (PAF). Each of these impact categories is described in Chapter 2. For EP and ecotoxicity a baseline value is included. The baseline considers effluent from the model site that would occur with no SCD installed. The baseline has no associated air emissions and is therefore not included in GWP or AP. Key items from the impact analysis are identified for each impact category in the following paragraphs. Global warming potential associated with straw bale barriers is 26 times greater than that for the mulch berm device. GWP for each SCD is divided into process-Type in Table 22 with values contributing at least 10% of GWP in bold. The large GWP for straw bales is attributed to landfill of the untreated wood waste. Creation of the timber for each SCD using wooden stakes is a sink for greenhouse gases and reduces overall GWP. Steel production for Type C silt fence is a significant source of GWP. GWP associated with combustion emissions from combined transport processes and diesel fuel production is low compared to GWP from production phases. The large amount of material required for straw bale installations increases transport trip required and results in higher GWP. Operation of diesel machinery for SCD cleanout has a much higher GWP than for truck transportation. 119 Global Warming Air [kg CO2-Equiv.] 10,000.0 9,000.0 8,000.0 7,000.0 6,000.0 5,000.0 4,000.0 3,000.0 2,000.0 1,000.0 0.0 3,427.26 Compost sock GWP 10,501.51 4,112.22 400.6 1,904.28 Mulch Berm Silt fence, type A Straw Bale Barrier Silt fence, type C Figure 77: Global warming potential (GWP) Table 22: Global warming potential (GWP) by process Global Warming Potential (Equiv. kg CO2) Bolded values are >10% of total Steel Timber Production Polypropylene Straw Mulching Trenching Use Cleanout Metal Disposal Plastic Timber Transport Transport Diesel Geotextile silt fence Type A -422.9 251.0 Type C 3,417.5 412.2 12.1 189.1 7.9 1,827.0 2.5 37.5 12.1 187.5 21.3 13.0 10.3 38.7 Compost sock -752.4 380.0 285.6 178.1 12.0 3,250.8 4.3 68.9 Straw bale -3,008.8 25.1 174.8 12,999.1 228.9 82.4 Mulch berm 163.8 182.5 54.4 120 Total 1,904.3 4,112.6 3,427.3 10,501.5 400.6 Acidification potential is similar for all devices except Type C silt fence. Table 23 lists AP by process. The table shows that high AP associated with Type C fence is from production of steel. For the mass of material created, production of polypropylene is comparable to diesel machinery operation. The large amount of wooden stakes required for straw bales installations show a significant amount of AP during production and disposal. Large amounts of bales and wood for straw bale installations increases transport trips resulting in greater AP. Acidification Air [kg H+ moles-Equiv.] 650 600 550 500 450 400 350 300 251.345 250 200 150 100 50 AP 107.831 131.976 656.081 336.188 Compost sock Mulch Berm Silt fence, type A Straw Bale Barrier Silt fence, type C Figure 78: Acidification potential (AP) Table 23: Acidification potential (AP) by process Acidification Potential (Equiv. kg H+ moles) Bolded values are >10% of total Geotextile silt fence Type A Type C Compost sock Straw bale Mulch berm 121 Steel 508.7 Timber 10.2 18.1 72.4 Production Polypropylene 42.6 70.0 64.5 Straw 6.6 Mulching 74.4 42.7 Trenching 3.2 3.2 Use Cleanout 49.3 48.9 46.4 45.6 47.6 Metal 6.3 Disposal Plastic 1.3 2.2 2.0 Timber 12.2 21.8 87.1 Transport Transport Diesel 1.1 4.4 1.8 97.9 12.1 12.5 22.3 26.7 17.6 Total 132.0 656.1 251.3 336.2 107.8 Eutrophication potential is much higher for straw bales than any other SCD and the baseline. The high EP for straw bale installations is due to the large amount of phosphate that leaches from straw material. Eutrophication is caused by large increases in nitrogen (N) or phosphorus (P). Almost all of EP for each device is due to the performance during the use phase. The use phase accounts for 98, 94, 96, 99 and 99 percent of total EP for the Type A, Type C, compost sock, straw bale and mulch berm devices, respectively. A small amount of EP is due to steel manufacture for the Type C device. 122 Eutrophication [kg N-Equiv.] 45 40 35 30 25 20 15 9.848 10 5 8.609 EP 7.739 8.444 49.946 7.718 BASELINE Mulch Berm Silt fence, type C Compost sock Silt fence, type A Straw Bale Barrier Figure 79: Eutrophication potential (EP) Ecotoxicity to water shows the most deviation between each device and the baseline. At least 99.8% of contributions to water ecotoxicity are from the use phases of each device. The mulch berm and compost sock show an approximate 90% reduction in ecotoxicity when compared to the baseline value. The straw bale barriers show a 55% reduction. Ecotoxicity reduction in the compost, mulch and straw devices is mostly due to the measured reduction in metals and nutrients from the site effluent. The reduction achieved by both silt fence types is relatively low because the devices were assumed to have no nutrient or metal retaining capabilities. The reduction shown for the silt fence material is due to the detention of runoff water and subsequent retention of nutrients and metal concentrations in the water. Ecotoxicity is 123 shown as the potential affected fraction (PAF). The PAF is area of the local environment affected for a certain time per kg of pollutant. Ecotoxcity Water [PAF m3 day/kg] 260,000 240,000 220,000 200,000 180,000 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 273,199.921 Ecotox Water 229,319.143 202,051.085 122,953.255 26,591.093 35,990.042 BASELINE Mulch Berm Silt fence, type C Compost sock Silt fence, type A Straw Bale Barrier Figure 80: Ecotoxicity to water GWP and AP for each life cycle phase are shown in Figure 81 and Figure 82, respectively. GWP and AP are mostly by steel production for Type C fence supports. GWP and AP from disposal are dominated by landfill of wood materials. 124 GWP (kg equiv. CO2) 14,000 12,000 10,000 8,000 6,000 4,000 2,000 0 -2,000 -4,000 Type A fence Straw bale Type C fence Mulch berm Compost sock Production Use Figure 81: GWP by phase Disposal AP (kg equiv. H+ ions) 700 600 Type A fence Type C fence Compost sock Straw bale Mulch berm 500 400 300 200 100 0 Production Use Figure 82: AP by phase Disposal 125 4.3.4 Discussion The first point of emphasis is that GWP and AP from overall production and disposal phases is much larger than transport and diesel machinery processes (i.e., mulching, clean out). This increases the confidence of the assessment because production and disposal processes included in this study are based on a large amount of data developed by PE INTERNATIONAL whereas distances for transport and power requirements for diesel machinery are only approximations and will vary from site to site. It also suggests greater potential for lowering environmental impact by focusing on reducing SCD production than developing more efficient transport and installation methods. The results of the life cycle analysis show that differences in GWP and AP for the use phase of SCDs are relatively small and only differ by the amount of diesel machinery operation required for cleanout. Differences in GWP and EP from production phases are due to the high energy requirement during steel production. The production of polypropylene on the scale covered in the model results in GWP and AP on par with operation of diesel equipment for mulch production. A large amount of GWP and AP is derived from landfill of wood waste. This is likely due to the fast degradation of wood that results in gas emissions. Gas collection and reuse is considered as part of the landfill of wood waste process and results in a power source within the product life cycle. Productive reuse of the gas material was not included in this model. Differences in EP are generally small between different SCDs except for the straw bale installation that generated a large EP due to high levels of phosphate leachate that were encountered during the laboratory testing. EP for the silt fence installations was similar to the compost sock and mulch berm even though no nutrient retention was assumed for the silt fences. This is explained by the relatively low amounts of nutrient concentrations (0.5 mg/L) assumed to 126 be in runoff water from the site. Additional testing with much higher nutrient concentrations (2 5 mg/L) should be performed to investigate EP during times of high nutrient loading, such as after grassing and/or fertilizing on site. Results for ecotoxicity are the best example of beneficial SCD installation. Each device showed a reduction in aquatic ecotoxicity when compared to the control model (baseline). EP due to nutrient loading showed little variation between baseline and SCD values (except straw bales) which suggests that the large variation in ecotoxicity values are due to metal retention. The compost and mulch material each showed very high metal retention capabilities under very short test duration. No metal retention was assumed for the geotextile fences, although retention of runoff water was assumed to reduce metal concentrations in the total effluent volume. 4.3.5 Overall Performance (Valuation) SCD performance varies significantly between the different impact categories analyzed so that the best environmental option is not easily identified. To determine the SCD with the overall lowest environmental impact, a valuation analysis is performed. The steps in the valuation analysis are as follows: 1. Weighting factors are applied to each impact category, 2. SCD performance in each impact category is normalized relative to the other devices, 3. The product of relative SCD performance and the impact weights is summed for each SCD to determine the overall environmental impact. As discussed in section 2.3.7 Valuation, developing proper weighting factors requires significant discussion and one of the available development methods. The Berrittella et al. (2007) report is recommended as a good example of developing weighting factors. The weighting factors assumed for this study are shown in Table 24. The weighting factors are 127 shown as fractions of the final goal, similar to an analytical hierarchy process (AHP) study. The final goal in this evaluation is to determine the SCD with the lowest environmental impact. The primary goal of SCDs is to reduce surface discharges to adjacent waterways, thereby reducing water pollution. This is included in the valuation by rating impact criteria associated with water quality twice as high (0.666) as those for air quality (0.333). Sub-criteria for air and water quality are assumed to be equally important. The total environmental impact is now described as the sum of effects from global warming, acidification, eutrophication, and aquatic toxicity. This relationship is expressed as: GWP + AP + EP + Toxicity = Environmental Impact 0.1665 + 0.1664 + 0.333 + 0.333 = 0.999 Table 24: Weighting factors for LCIA valuation. Start: Criteria: Sub- criteria: Environmental Impact (EI) 1.000 Air Quality Water Quality 0.333 0.666 GWP AP EP Toxicity 0.1665 0.1665 0.333 0.333 Relative performance of SCDs in each impact category is expressed as the fraction of impact from each SCD to the total impact from all the SCDs analyzed. Relative performance values are shown in Table 25 for air quality and Table 26 for water quality. An example of relative performance calculation for GWP of compost sock is shown below. - = = 0 + 1904.28 + 3427.26 4112.22 + 3427.26 + 400.6 + 10501.51 = 0.168 128 Table 25: Relative SCD performance, air quality. Baseline Type A Fence Type C Fence Compost Sock Mulch Berm Straw Bale GWP kg CO2Eq % of max 0 0.000 1904.28 0.094 4112.22 0.202 3427.26 0.168 400.6 0.020 10501.51 0.516 AP kg H+ moles-Eq % of max 0 0.000 131.976 0.089 656.081 0.442 251.345 0.169 107.831 0.073 336.188 0.227 Table 26: Relative SCD performance, water quality. Baseline Type A Fence Type C Fence Compost Sock Mulch Berm Straw Bale EP kg N-Eq % of max 9.848 8.444 7.718 8.609 7.739 49.946 0.107 0.091 0.084 0.093 0.084 0.541 Toxicity PAF m3 day/kg % of max 273199.9 0.307 229319.1 0.258 202051.1 0.227 26591.09 0.030 35990.04 0.040 122953.3 0.138 Relative SCD performance is multiplied by the corresponding impact category weighting factor to determine the final overall environmental impact. An example for compost sock is included below. Results of the valuation are shown in Figure 83. = - + - + - + - = 0.1665 0.168 + 0.1665 0.169 + 0.333 0.093 + 0.333 0.030 = 0.097 129 Where EI is the environmental impact for compost sock, V is the weight factor for each impact category, and I is the relative performance of compost sock in each impact category. Relative Environmental Impact Relative Environmental Impact 0.350 0.057 0.097 0.138 0.147 0.211 Mulch Berm 18" Compost Baseline/No Type A Silt Type C Silt Straw Bale Sock Device Fence Fence Figure 83: Results of LCIA valuation. SCDs are ranked from lowest to highest relative environmental impact. Results of the valuation indicate that for the assumed site conditions, the mulch berm has the lowest environmental impact, followed by the compost sock and then no SCD device. For the assumed weighting factors and the field and lab testing performed, the overall impact of silt fence installation is higher than no SCD installation. An additional valuation was performed using the relative performance of SCD devices for TSS removal and turbidity reduction. The added valuation was performed to compare the typical decision making criteria of lower TSS and turbidity to results from the life-cycle study. For the SCD performance valuation, TSS removal and turbidity reduction are considered to be equally important, as shown by weighting factors in Table 27. Relative SCD performance for 130 TSS removal and turbidity reduction are shown in Table 28. The baseline (no SCD installation) has no reduction potential. Results of the SCD performance valuation are shown in Figure 84. Results indicate that silt fence is the best option for TSS removal and turbidity reduction, but are only marginally better than each of the other SCD options. Table 27: Weighting factors for SCD performance valuation. Start: Criteria: SCD Performance 1.000 TSS Removal Turbidity Reduction 0.500 0.500 Table 28: Relative SCD performance, TSS removal and turbidity reduction. Baseline Type A Fence Type C Fence Compost Sock Mulch Berm Straw Bale TSS Removal Reduction (%) % of max 0 0.00 98.4 0.21 97.6 0.21 92.9 0.20 95 0.20 91.2 0.19 Turbidity Reduction Reduction (%) % of max 0 0.00 92.3 0.24 92.8 0.24 64.9 0.17 73.8 0.19 61.9 0.16 131 0.223 TSS/Turbidity Reduction TSS Removal & Turbidity Reduction 0.223 0.196 0.182 0.176 0.000 Type A Silt Type C Silt Mulch Berm 18" Compost Straw Bale Baseline/No Fence Fence Sock Device Figure 84: Results of SCD performance valuation. SCDs are ranked from best (highest) to worst (lowest) relative performance. 4.3.5 Implications Results from the study suggest that straw bale installations may degrade downstream water systems by adding a large amount of phosphate to water leaving construction sites and thus increasing the potential for eutrophication. In addition, although timber production is a sink for greenhouse gasses, the large amount of wood waste from wooden stakes creates a much larger source for harmful emissions. The assumed duration on site of one year is about the usable lifetime of an untreated wooded stake and reuse of the wood stakes to avoid landfill is unlikely. However, emissions from degradation of the wood waste may be used for power generation in order to lower the apparent environmental impact during modeling. Regardless, the poor performance of straw bales as sediment barriers is noted by many state DOTs that currently no not allow bales for use as perimeter barriers. Another implication from the study is that high flow (Type A) geotextile silt fence performs better than low flow (Type C) silt fence. During field testing the devices, the Type A 132 fence resulted in overall lower sediment downstream of the installation. Type A fence shows slightly more EP and aquatic toxicity due to the higher retention of runoff water of the Type C fence. However, GWP and AP associated with the Type C fence are much higher than those of the Type A fence due to the energy intensive production of steel. The steel support is required for additional support of silt fence, however; during field testing, the Type A fence performed well, if not better than the Type C fence (first test of Type C fence resulted in undercutting failure). Although polypropylene is the result of the energy intensive process of cracking of hydrocarbons, cumulative GWP and AP associated with the material is on the scale of GWP and AP from operating equipment for mulching and trenching. Although, the use of steel stakes and supporting wire mesh adds environmental load, the impact of polypropylene is not large enough to discourage its use as SCD material. A single field test of each SCD with total maximum test duration of 90 minutes was performed for this research. Mulch berms and compost socks performed very well during testing; however, the performance of these devices on the scale of the model site duration (1 year) was not measured. All installations modeled are assumed to stay in place without replacement for the entire duration; only maintenance to clean out the SCDs was modeled. Still, the size and weight of compost socks suggests that they would stay in place for long durations. Additionally, the mulch berm tested was much smaller than typical brush barriers formed on sites from slash material during site clearing stage. Results from the LCIA valuation indicate that the mulch berm is the SCD with the lowest overall environmental impact. Results from the SCD performance indicate that silt fence is the best option to reduce TSS and turbidity. The difference in the results underlines the ability of LCA to identify impacts that were not previously considered. The difference in performance 133 between the worst performing SCD (straw bales) to the best performing SCD (mulch berm) is approximately 83%. The largest difference in performance from the SCD valuation is only 21%, from silt fence to straw bales. The differences suggest that the SCD valuation based on only TSS and turbidity reduction is slight. The valuation based on life-cycle performance is much greater, suggesting the best-option is far better than the worst option. 4.3.6 Limitations Life cycle analysis provides a holistic approach to comparing the environmental impacts of processes; however, the analysis is still based on a model of a real system and simplification errors are unavoidable. Multiple modeling assumptions and other general shortcomings of the analysis are discussed in this section. Problems associated with the field and laboratory testing are discussed in previous sections. In this model, steel fence posts are assumed as steel sections and wire mesh is assumed as steel wire rod. A power requirement for weaving textile in Indonesia is assumed as applicable to weaving polypropylene geotextile in the southeast United States. Processes refined in Europe for production of polypropylene fibers and disposal of wood waste, metal and plastic and production of timber in Germany are assumed to be applicable in Georgia. Using foreign processes is generally okay as production phases are not expected to vary spatially as much as power generation; diesel refining processes specific to the United States were used in the study. Inventories for common processes like steel and plastic production are based on a large amount of industry data and analysis. For this model, installation and maintenance requirements are only approximations based on power requirements. Accuracy of the model with increase with additional data on machine emissions, fuel requirement and actual machine hours required for installation and maintenance. 134 The model developed for this study assumed a 1,000 meter (0.62 mile) long installation operating at a site for 1-year. The site was assumed to be located 10 kilometers (6.2 miles) from SCD supply plants and 25 kilometers (15.5 miles) from landfill facilities. Distances from supply plants may vary for each SCD. Total precipitation on site was assumed as the yearly average for Macon, Georgia (~46"). The rainfall was assumed to occur over 11.5 storms each with the intensity of a 10-year, 6-hour storm. This was the test condition assumed during the field testing of the SCDs, actual performance of the SCD devices is expected to vary significantly with different storm intensities. Nutrient retention was not considered for silt fence in this study. Slight differences in nutrient retention are realized for the silt fence due to the water retention capabilities. Additional studies of nutrient retention for all devices in the field will increase LCA accuracy. The LCIA and SCD performance valuations provide insight to SCD selection; however, the valuations are based on assumed weighting factors. Additionally, at the time of this report, it is not clear how the TRACI aquatic eco-toxicity impact category quantifies affects from downstream TSS loads and turbidity. 135 CONCLUSION 5.1 Sediment Control Device Performance This study was performed for the Georgia Department of Transportation (GDOT) in order to better understand the environmental impacts associated with sediment control devices currently employed on transportation projects. In this study, current and past methods for testing sediment control devices (SCDs) in the field and laboratory and procedures for conducting a life cycle assessment were reviewed. Extensive field testing was performed in accordance with ASTM D7351. TSS and turbidity reduction performance in the field tests is summarized in Table 29. The table ranks the devices in two columns according to performance. The tests indicate that geotextile silt fence performed the best in both TSS reduction and turbidity reduction. TSS reduction only varied by 7.2% from Type A fence to straw bales. Turbidity varied by 43.6% from the best, Type C fence, to the worst, straw bales. Table 29: SCD TSS and turbidity removal performance. TSS Reduction Device: Removal Efficiency (%) Type A Fence: 98.4 Type C Fence: 97.6 Mulch Berm: 95.0 18" Compost Sock: 92.9 Straw Bales: 91.2 Turbidity Reduction Device: Removal Efficiency (%) Type C Fence: 92.8 Type A Fence: 92.3 Mulch Berm: 73.8 18" Compost Sock: 64.9 Straw Bales: 49.2 A combination of laboratory filter testing was performed to determine the capability of SCD nutrient and metal retention capabilities. Results, summarized in Table 30, indicate that 136 retention of nitrogen in mulch berms is significantly higher than in straw and compost and that removal of phosphorous is generally low. Straw bales are a significant source of additional phosphorous. Table 30: Nutrient retention results. Total Nitrogen Reduction Device: Removal Efficiency (%) Mulch: 56.1 Straw: 32.5 Compost: 6.4 Phosphorous Reduction Device: Removal Efficiency (%) Compost: 8.2 Mulch: 0.0 Straw: -811.6 Table 31 ranks compost, mulch, straw according to metal retention capability. Retention in compost and mulch is high. Retention in straw is moderate. Table 31: Metal retention results. Copper (Cu) Reduction Device: Removal Efficiency (%) Compost:93.8 Mulch: 89.0 Straw: 47.0 Zinc (Zn) Reduction Device: Removal Efficiency (%) Compost: 85.9 Mulch: 79.6 Straw: 26.0 Lead (Pb) Reduction Device: Removal Efficiency (%) Compost: 91.7 Mulch: 87.2 Straw: 47.6 Overall results of the SCDs indicate that silt fence is the best option for reducing TSS and turbidity in downstream water samples. Mulch is the best option for reducing downstream nutrients. Compost is the best option for retaining metals in runoff water. Taken without performing a LCA, the SCD performance testing suggests that when nutrient and metal concentrations exist in runoff streams, mulch and compost are good modifiers to be used in 137 combination with silt fence. A valuation of the SCD performance indicates that silt fence are the best option to reduce TSS and turbidity, but are only slightly better than other SCD options. 5.2 Life Cycle Analysis of Sediment Control Devices The field and lab tests were combined to create a use phase process for each SCD. The use phase processes were combined in a life cycle inventory with previously created production and disposal processes included in the GaBi 6.0 professional life cycle database. Additional processes were approximated based on power requirements found in literature. The processes were linked to form a life cycle plan of each SCD. The life cycle model included performance of the SCDs on a 1,000 meter long installation for 1 year under assumed runoff conditions. A life cycle impact analysis was performed using GaBi 6.0 software and USEPA TRACI methodology. An LCIA valuation was performed using assumed weighting factors to determine the overall most environmentally friendly SCD option. Results of the LCA indicate: 1. Straw bale installations significantly increase eutrophication potential in downstream water systems due to high levels of phosphate present in the straw bales, 2. Production of steel sections and wire mesh for support of low-flow (Type C) silt fence result in large increases in global warming and acidification potential, 3. Good performance of Type A silt fence suggests it is a good alternative to Type C fence in extreme conditions, 4. Overall low global warming and acidification potentials as well as low aquatic toxicity levels attributable to mulch berms suggests their use as an alternative SCD to geotextile silt fence is favorable, 5. A preliminary LCIA valuation with assumed impact weighting factors indicates that mulch berm is the SCD with the lowest overall environmental impact. 138 5.3 Impact/Recommendations Currently, selection of SCD devices is based on TSS removal and turbidity reduction. The SCD valuation indicates that although silt fence is favorable, it is not significantly better than other SCD devices. The LCIA valuation adds impacts to global warming, acidification, eutrophication, and toxicity that include production and disposal phases. The major re-ordering of SCD performance between the two valuations reveals the significance of previously unconsidered environmental impacts on SCD selection. The following recommendations are derived based on results from this study: Mulch berms should be considered favorable for use as perimeter SCDs, Evaluation of new and existing sediment control devices should include a life- cycle analysis to compare overall environmental performance, and Results from general LCA studies can be tailored to individual construction sites using valuation processes and weighted impact factors determined at the local level. Life-cycle analysis is beneficial but also costly. As data is accumulated from additional SCD performance testing, as impact categories are refined to include affects from downstream TSS and turbidity, and as energy and material cost for SCD materials are better understood, increasingly accurate LCA studies can be determined. LCA studies for every SCD installation would require significant time and may be too costly to sustain for a long period of time. However, general LCA studies to describe SCD performance in different conditions are beneficial to describe relative performance of devices. Results from the general studies can be modified using weighting factors that reflect the local site conditions (i.e., vicinity to vulnerable waterway, location in urban area with high air quality standards). 139 5.4 Future Work On the basis of the results of this study, the following recommendations are presented for future work: 1. Nutrient and metal retention of SCDs should be studied with advanced laboratory procedures and in-field conditions. 2. Field and large-scale testing of SCDs for TSS retention and turbidity reduction should be studied for various sediment and water loads. Additional large-scale testing of SCDs should be done to study the effect of sediment grain size on filtering performance. 3. Advanced studies should be done to identify the mode of filtration of SCDs for sediment, nutrient, and metal reductions. 4. Additional LCAs should be performed to investigate the costs and benefits for recycling SCD components. The feasibility of SCD recycling should be investigated in the field. 5. Additional studies should determine appropriate weighting factors to be used for various on-site conditions. 140 APPENDIX A: DOT SUMMARY Sources for DOT search summarized below, general form of data is: [State] [DOT website] [Name of E&SC manual] [Online access] Alaska www.dot.state.ak.us Alaska Storm Water Guide. Alaska Department of Environmental Conservation. http://dec.alaska.gov/water/wnpspc/stormwater/docs/AKSWGuide.pdf Arizona http://www.azdot.gov/ ADOT Erosion and Pollution Control Manual for Highway Design and Construction http://www.azdot.gov/Highways/Roadway_Engineering/Roadside_Development/Resources.asp Arkansas http://www.arkansashighways.com/ 2004 Erosion and Sediment Control Design and Construction Manual http://www.arkansashighways.com/Construc/2004_E&S_Control_Manual/1104%20E%20SC%20MANUAL%20FINAL.pdf California http://www.dot.ca.gov/ Caltrans Storm Water Quality Handbooks Construction Site Best Management Practices Manual Section 1 March 1, 2003 http://www.dot.ca.gov/hq/construc/stormwater/CSBMPM_303_Final.pdf 141 Colorado http://www.coloradodot.info/ CDOT Erosion Control and Stormwater Quality Field Guide http://www.coloradodot.info/programs/environmental/waterquality/documents/CDOT%20Pocket% 20Guide%20122211.pdf Connecticut http://www.ct.gov/dot/site/default.asp 2002 Connecticut Guidelines for Soil Erosion and Sediment Control http://ahhowland.com/regulations/state-of-ct/ct-dep/2002-ct-guidelines-for-sediment-and-erosioncontrol.pdf Delaware http://www.deldot.gov/ E&S Field guide http://www.deldot.gov/stormwater/pdfs/EandS_fieldguide/III+PerimeterControlsDiversionAndInlet Protection.pdf Florida http://www.dot.state.fl.us/ State of Florida Erosion and Sediment Control Designer and Reviewer Manual, June 2007 http://www.dot.state.fl.us/rddesign/dr/files/Erosion-Sediment-Control.pdf Georgia http://www.dot.ga.gov/Pages/default.aspx Manual for Erosion and Sediment Control in Georgia Hawaii http://hidot.hawaii.gov/ HiDOT - Construction Best Management Practices Field Manual, January 2008 http://www.coralreef.gov/transportation/constructionmanual_022708.pdf 142 Idaho http://itd.idaho.gov/ Best Management Practices, Erosion and Sediment Control, ITD, August 2008 http://itd.idaho.gov/manuals/Online_Manuals/Current_Manuals/BMP/BMP%20Manual.pdf Illinois http://www.dot.state.il.us/ IDOT Erosion and Sediment Control Field Guide for Construction Inspection, July 1, 2010 http://www.dot.il.gov/desenv/environmental/idot%20field%20guide.pdf Indiana http://www.in.gov/indot/ 2013 Indiana Design Manual http://www.in.gov/indot/design_manual/files/Ch205_2013.pdf Iowa http://www.iowadot.gov/ Iowa DOT Erosion & Sediment Control Field Guide, April 2012 http://www.iowadot.gov/construction/earthwork_erosion/Erosion_Sediment_Control_Field_Guide. pdf Kansas http://www.ksdot.org/ KDOT Temporary Erosion Control Manual, January 2007 http://www.ksdot.org/burconsmain/Connections/ecm.pdf Kentucky http://transportation.ky.gov/Pages/default.aspx Kentucky BMPs for Controlling Erosion, Sediment, and Pollutant runoff from construction Sites http://transportation.ky.gov/environmentalanalysis/environmental%20resources/ky%20bmp%20m anual%20section%201.pdf Louisiana http://www.dotd.state.la.us/ http://www.dotd.la.gov/highways/construction/lab/pdf/embankment/temporary%20erosion%20con trol.pdf 143 Maine http://www.maine.gov/mdot/ Erosion and Sediment Control BMP, revised October 2012 http://www.maine.gov/dep/land/erosion/escbmps/ Maryland http://www.mdot.maryland.gov/ 2011 Maryland Standards and Specifications for Soil Erosion and Sediment Control, December 2011 http://www.mde.maryland.gov/programs/Water/StormwaterManagementProgram/SoilErosionand SedimentControl/Documents/2011%20MD%20Standard%20and%20Specifications%20for%20So il%20Erosion%20and%20Sediment%20Control.pdf Massachusetts http://www.massdot.state.ma.us/ Massachusetts Erosion and Sediment Control Guidelines for Urban and Suburban Areas: A Guide for Planners, Designers, and Municipal Officials, reprint May 2003 http://www.mass.gov/dep/water/laws/policies.htm#storm Michigan http://michigan.gov/mdot/ MDOT Soil Erosion and Sedimentation Control Manual, April 2006 http://www.michigan.gov/documents/2006_SESC_Manual_165226_7.pdf Minnesota http://www.dot.state.mn.us/ MnDOT Erosoin and Sediment Control Pocketbook Guide, June 2009 http://www.dot.state.mn.us/environment/pdf/erosion-sediment-control-handbook.pdf Mississippi http://mdot.ms.gov/portal/home.aspx Field Manual for Erosion and Sediment Control in Mississippi, 2nd edition 2005 http://www.deq.state.ms.us/MDEQ.nsf/pdf/NPS_Field_Manual_For_Erosion_And_Sediment_Con trol_Version_2/$File/NPS_FieldManualV2.pdf?OpenElement 144 Missouri http://www.modot.org/ MoDOT Stormwater Pollution Prevention Plan, June 2012 http://www.modot.org/business/contractor_resources/documents/MoDOTSWPPPJune2012.pdf Montana http://www.mdt.mt.gov/ Erosion and sediment control BMPs: Reference Manual, March 2003 (FHWA/MT-03-006/8165) http://www.mdt.mt.gov/research/projects/env/erosion.shtml Nebraska http://www.dor.state.ne.us/ Nebraska Department of Roads Construction Stormwater BMPs http://www.dor.state.ne.us/environment/guides/Const-Strmwtr-Pocket%20Guide.pdf Nevada http://www.nevadadot.com/ Nevada DOT Storm Water Quality Manuals: Construction Site BMPs Manual, January 2006 http://www.nevadadot.com/uploadedFiles/NDOT/About_NDOT/NDOT_Divisions/Engineering/Hyd raulics/2006_Storm_Water_Quality_BMP_Manual.pdf New Hampshire http://www.nh.gov/dot/ New Hampshire Stormwater Manual: Volume 3, Erosion and Sediment Controls During Construction, December 2008 http://des.nh.gov/organization/commissioner/pip/publications/wd/documents/wd-08-20c.pdf New Jersey http://www.state.nj.us/transportation/ New Jersey DOT Soil Erosion and Sediment Control Standards, 2008 http://www.state.nj.us/transportation/eng/documents/SESC/pdf/SESCStandards2008.pdf 145 New Mexico http://www.dot.state.nm.us/content/nmdot/en.html National Pollutant Discharge Elimination System Manual: Storm Water Management Guidelines for Construction and Industrial Activities, August 2012 http://www.dot.state.nm.us/content/dam/nmdot/Infrastructure/NPDESM.pdf New York https://www.dot.ny.gov/index?nd=nysdot New York State Standards and Specifications for Erosion and Sediment Control, August, 2005 http://www.dec.ny.gov/chemical/29066.html North Carolina http://www.ncdot.gov/ Erosion and Sediment control Planning and Design Manual, June 2006, updated March 2009 http://portal.ncdenr.org/web/lr/publications#espubs North Dakota http://www.dot.nd.gov/ NDDOT Erosion and Sediment Control Handbook, June 2004 http://www.dot.nd.gov/manuals/environmental/escm/escmfinal.pdf Ohio http://www.dot.state.oh.us/pages/home.aspx Ohio DOT Handbook for Sediment and Erosion Control, February 2000 http://www.dot.state.oh.us/Divisions/ConstructionMgt/Admin/Manuals/Erosion%20Control.pdf Oklahoma http://www.okladot.state.ok.us/ Oregon http://www.oregon.gov/odot/pages/index.aspx Erosion and Sediment Control Manual, April 2005 http://www.deq.state.or.us/wq/stormwater/docs/escmanual/manual.pdf 146 Pennsylvania http://www.dot.state.pa.us/ PennDOT Maintenance Field Reference for Erosion and Sediment Controls, 2003 ftp://ftp.dot.state.pa.us/public/PubsForms/Publications/PUB%20464.pdf Rhode Island http://www.dot.state.ri.us/ Rhode Island Soil Erosion and Sediment Control Handbook, 1989 http://www.dot.ri.gov/documents/enviro/stormwater/Soil_Erosion_Sediment_Control_Handbook.p df South Carolina http://www.dot.state.sc.us/ http://www.scdot.org/doing/stormwater_Erosion.aspx South Dakota http://www.sddot.com/ http://sddot.com/business/design/docs/wqep/section3.pdf Tennessee http://www.tdot.state.tn.us/ Tennessee Erosion & Sediment Control Handbook, August 2012 http://www.tnepsc.org/TDEC_EandS_Handbook_2012_Edition4/TDEC%20EandS%20Handbook %204th%20Edition.pdf Texas http://www.txdot.gov/ Texas DOT Storm Water Field Inspector's Guide, June 2004 http://ftp.dot.state.tx.us/pub/txdot-info/library/pubs/bus/natural/inspectors_guide.pdf Utah http://www.udot.utah.gov/main/f?p=100:6:0::::V,T:,1 Utah DOT Erosion and Sediment Control Field Guide, July 2010 http://www.udot.utah.gov/main/uconowner.gf?n=15220806279436191 147 Vermont http://www.aot.state.vt.us/ The Vermont Standards and Specifications for Erosion Prevention and Sediment Control, 2006 http://www.anr.state.vt.us/dec/waterq/stormwater/docs/construction/sw_vt_standards_and_specifi cations_2006_updated_2_20_2008.pdf Virginia http://www.virginiadot.org/ Virginia Erosion and Sediment Control Handbook, 1992 http://www.dcr.virginia.gov/stormwater_management/e_and_s-ftp.shtml Washington http://www.wsdot.wa.gov/ Highway Runoff Manual, Chapter 6: Temporary Erosion and Sediment Control Design Guidelines and Process, November 2011 http://www.wsdot.wa.gov/publications/manuals/fulltext/M31-16/chapter6.pdf West Virginia http://www.transportation.wv.gov/Pages/default.aspx WV DOT Division of Highways Erosion and Sediment Control Manual, March 2003 http://www.transportation.wv.gov/highways/engineering/files/EROSION/Erosion2003.pdf Wisconsin http://www.dot.wisconsin.gov/ Wisconsin Erosion Control Product Acceptability List, July 2012 http://www.dot.wisconsin.gov/business/engrserv/docs/pal.pdf Wyoming http://www.dot.state.wy.us/wydot Wyoming DOT Pollution controls and BMPs for Storm Water During Construction http://www.dot.state.wy.us/files/content/sites/wydot/files/shared/Construction/WYDOT-fieldguide,4-6-11%5B1%5D.pdf 148 APPENDIX B: TSS/TDS RESULTS Measured total suspended solids (TSS) and conductivity and corresponding total dissolved solids (TDS) values shown for each sample collected during ASTM D 7351 tests. Conductivity was not measured and therefore TDS not calculated for the initial (failed) test of Type C silt fence, test number 2. Test numbers are as described in Table 15: Summary of ASTM D7351 testing. Values not measured are indicated by NM. Table B-1: Measured TSS, test 1. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Trial 1 TSS (mg/L) 1429.3 2137.3 686.2 2625.5 734.3 377.5 89.7 70.0 65.5 33.8 36.5 47.0 194.5 42.0 37.0 35.5 Trial 2 TSS (mg/L) 1437.0 2144.5 741.7 2436.4 734.9 354.2 103.8 99.4 73.4 46.8 39.5 45.2 36.7 32.4 29.6 38.1 Average TSS (mg/L) 1433.14 2140.90 713.93 2530.95 734.64 365.85 96.71 84.70 69.43 40.32 38.00 46.07 115.60 37.20 33.30 36.80 149 Table B-2: Measured conductivity/TDS, test 1. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 Trial 1 Conductivity (us/cm) Trial 2 Conductivity (us/cm) Average Conductivity (us/cm) 107.83 72.04 65.47 64.35 65.35 52.18 84.46 88.31 90.85 88.15 83.85 106.15 99.23 98.23 94.62 99.00 94.62 105.46 93.46 94.69 96.23 99.08 96.54 102.46 100.77 89.46 90.62 97.08 94.46 93.31 91.77 91.08 101.22 88.75 79.47 79.52 80.79 75.63 90.50 95.38 95.81 88.81 87.23 101.62 96.85 95.77 93.19 95.04 TDS (mg/L) y=0.6798x x=cond. 68.81 60.33 54.02 54.06 54.92 51.41 61.52 64.84 65.13 60.37 59.30 69.08 65.84 65.10 63.35 64.61 Table B-3: Measured TSS, test 2. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 Trial 1 TSS (mg/L) 6035.0 7547.5 3237.5 9282.5 12100.0 2875.0 14747.5 1055.0 15.0 Trial 2 TSS (mg/L) 6767.5 12265.0 5537.5 10137.5 12962.5 3172.5 15407.5 1485.0 145.0 Average TSS (mg/L) 6401.25 9906.25 4387.50 9710.00 12531.25 3023.75 15077.50 1270.00 80.00 150 Table B-4: Measured TSS, test 2B. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Trial 1 TSS (mg/L) 4722.6 768.8 1572.6 2802.1 1944.6 930.2 147.6 100.1 57.4 58.0 52.8 58.8 44.1 73.6 58.5 76.3 26.7 26.2 Trial 2 TSS (mg/L) 5100.3 883.9 1842.5 3524.1 2252.8 1053.9 177.7 122.8 61.3 62.3 58.8 60.8 56.6 80.2 60.3 78.4 30.3 24.6 Average TSS (mg/L) 4911.43 826.33 1707.55 3163.10 2098.68 992.03 162.60 111.48 59.33 60.15 55.78 59.82 50.35 76.90 59.38 77.38 28.50 25.45 Table B-5: Measured conductivity/TDS, test 2B. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 Trial 1 Conductivity (us/cm) 86.31 94.77 99.77 110.92 101.38 111.77 71.08 96.54 91.54 110.62 107.92 112.92 119.69 125.23 120.54 125.15 114.69 Trial 2 Conductivity (us/cm) 112.00 111.31 110.00 110.92 99.62 98.15 101.54 76.15 106.46 112.85 108.92 115.31 117.46 117.08 117.00 115.38 110.77 Average Conductivity (us/cm) 99.15 103.04 104.88 110.92 100.50 104.96 86.31 86.35 99.00 111.73 108.42 114.12 118.58 121.15 118.77 120.27 112.73 TDS (mg/L) y=0.6798x x=cond. 67.40 70.05 71.30 75.41 68.32 71.35 58.67 58.70 67.30 75.95 73.71 77.58 80.61 82.36 80.74 81.76 76.63 151 90 114.31 105.85 110.08 74.83 Table B-6: Measured TSS, test 3. Elapsed Time (min) 5 10 15 20 25 30 Trial 1 TSS (mg/L) 3101.2 2492.7 7585.9 12589.6 12812.5 6854.5 Trial 2 TSS (mg/L) 2883.6 2329.7 8006.6 12305.2 11633.8 6420.8 Average TSS (mg/L) 2992.38 2411.20 7796.23 12447.40 12223.15 6637.63 Table B-7: Measured conductivity/TDS, test 3. Elapsed Time (min) 5 10 15 20 25 30 Trial 1 Conductivity (us/cm) Trial 2 Conductivity (us/cm) Average Conductivity (us/cm) 176.00 135.15 113.54 139.75 116.85 110.54 122.00 111.43 98.07 81.64 95.71 65.94 149.00 123.29 105.80 110.70 106.28 88.24 TDS (mg/L) y=0.6798x x=cond. 101.29 83.81 71.93 75.25 72.25 59.98 Table B-8: Measured TSS, test 4. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 Trial 1 TSS (mg/L) 4114.1 4249.4 7968.3 8652.1 7815.3 6843.5 941.4 610.5 455.7 412.0 Trial 2 TSS (mg/L) 4180.7 4451.7 7908.2 8697.4 7790.4 6725.6 907.2 550.9 398.4 219.8 Average TSS (mg/L) 4147.38 4350.57 7938.21 8674.73 7802.83 6784.55 924.28 580.70 427.10 315.90 152 Table B-9: Measured conductivity/TDS, test 4. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 Trial 1 Conductivity (us/cm) Trial 2 Conductivity (us/cm) Average Conductivity (us/cm) 142.92 130.23 115.15 125.69 144.92 125.17 168.09 166.18 138.58 189.36 142.92 130.23 115.15 125.69 144.92 125.17 168.09 166.18 138.58 189.36 142.92 130.23 115.15 125.69 144.92 125.17 168.09 166.18 138.58 189.36 TDS (mg/L) y=0.6798x x=cond. 97.16 88.53 78.28 85.45 98.51 85.09 114.27 112.97 94.21 128.73 Table B-10: Measured TSS, test 5. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 Trial 1 TSS (mg/L) 3662.7 4843.5 3205.7 3394.0 2787.5 4595.6 1798.8 1034.7 357.35 Trial 2 TSS (mg/L) 3755.75 4869.5 3003 3503.35 2801.2 4643.9 1750.75 1056.6 351.55 Average TSS (mg/L) 3709.23 4856.50 3104.35 3448.68 2794.35 4619.73 1774.78 1045.63 354.45 153 Table B-11: Measured conductivity/TDS, test 5. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 Trial 1 Conductivity (us/cm) Trial 2 Conductivity (us/cm) Average Conductivity (us/cm) 129.33 130.42 115.08 124.42 113.33 129.33 128.50 138.33 147.33 142.50 143.00 99.00 131.42 131.92 132.75 156.18 139.67 140.08 135.92 136.71 107.04 127.92 122.63 131.04 142.34 139.00 143.71 TDS (mg/L) y=0.6798x x=cond. 92.40 92.93 72.77 86.96 83.36 89.08 96.76 94.49 97.69 Table B-12: Measured TSS, test 6. Elapsed Time (min) 5 10 15 17.5 20 25 30 35 40 45 50 55 60 65 70 75 80 Trial 1 TSS (mg/L) 4743.5 4990.0 4800.0 6231.7 8908.3 6845.0 5463.3 4850.0 7183.3 4913.3 1951.7 843.3 1008.3 825.8 1150.8 1172.5 1151.7 Trial 2 TSS (mg/L) 4731.67 4990 4780 6010 8590 6870 5457.5 5510 6462.5 4860 1730 1003.75 746.67 675 998.33 1038.33 998.33 Average TSS (mg/L) 4737.58 4990.00 4790.00 6120.83 8749.17 6857.50 5460.42 5180.00 6822.92 4886.67 1840.83 923.54 877.50 750.42 1074.58 1105.42 1075.00 154 Table B-13: Measured conductivity/TDS, test 6. Elapsed Time (min) 5 10 15 17.5 20 25 30 35 40 45 50 55 60 65 70 75 80 Trial 1 Conductivity (us/cm) Trial 2 Conductivity (us/cm) Average Conductivity (us/cm) 559.29 NM 559.29 321.39 NM 321.39 283.44 NM 283.44 109.60 NM 109.60 94.35 NM 94.35 66.65 NM 66.65 116.76 NM 116.76 105.24 NM 105.24 274.71 NM 274.71 222.23 NM 222.23 236.33 NM 236.33 211.05 NM 211.05 294.00 NM 294.00 261.60 NM 261.60 247.97 NM 247.97 335.23 NM 335.23 282.17 NM 282.17 TDS (mg/L) y=0.6798x x=cond. 380.20 218.48 192.68 74.51 64.14 45.31 79.37 71.54 186.75 151.07 160.65 143.47 199.86 177.84 168.57 227.89 191.82 155 APPENDIX C: TURBIDITY RESULTS Measured turbidity vales for samples collected downstream of SCD installation are shown in following tables. Test numbers are as described in Table 15: Summary of ASTM D7351 testing. Values not measured are indicated by NM. Table C-1: Measured turbidity downstream, test 1. Elapsed Time (min) NTU 1 TRIAL 1 NTU 2 NTU 3 5 442 463 455 10 475 500 494 15 276 307 295 20 537 531 525 25 267 276 273 30 390 377 386 35 139 140 134 40 122 124 125 45 85 85 84 50 60 57 62 55 66 63 66 60 67 71 73 65 71 77 74 70 61 62 65 75 66 65 69 80 70 71 74 TRIAL 2 NTU 1 NTU 2 NTU 3 331 319 309 458 465 465 312 285 283 528 524 513 330 351 339 417 392 403 139 132 140 135 122 127 86 85 89 66 61 72 64 63 59 77 71 74 79 75 75 65 65 64 64 65 65 71 76 74 156 Table C-2: Measured turbidity downstream, test 2. Elapsed Time (min) 5 10 15 20 25 30 35 40 NTU 1 402 424 410 417 564 224 792 168 TRIAL 1 NTU 2 409 444 338 397 563 215 784 162 NTU 3 387 446 330 392 560 215 778 150 TRIAL 2 NTU 1 NTU 2 NTU 3 379 372 400 342 328 391 430 421 475 514 503 462 549 537 569 244 285 243 667 746 659 157 157 175 Table C-3: Measured turbidity downstream, test 2b. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 NTU 1 632 416 452 704 392 259 180 151 109 108 99.6 92 95.5 136 100 109 59.7 54.8 TRIAL 1 NTU 2 590 417 457 720 423 295 197 158 108 112 101 89.9 96.6 129 101 106 58.7 57 NTU 3 590 401 450 737 403 268 200 154 109 110 103 96 98 137 98.4 111 62.5 60.1 TRIAL 2 NTU 1 NTU 2 NTU 3 420 416 401 288 283 301 350 365 358 544 570 584 515 565 519 260 264 264 NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM NM 157 Table C-4: Measured turbidity downstream, test 3. Elapsed Time (min) 5 10 15 20 25 30 NTU 1 233 403 524 624 566 377 TRIAL 1 NTU 2 259 395 506 638 582 382 NTU 3 260 399 483 660 615 395 TRIAL 2 NTU 1 NTU 2 NTU 3 289 296 305 513 514 555 530 581 504 599 652 637 460 489 503 465 487 469 Table C-5: Measured turbidity downstream, test 4. Elapsed Time (min) 5 10 15 20 25 30 35 40 45 50 NTU 1 617 744 732 836 621 763 489 285 263 524 TRIAL 1 NTU 2 585 647 737 814 586 766 515 276 270 524 NTU 3 582 719 742 785 653 750 467 277 258 520 TRIAL 2 NTU 1 NTU 2 NTU 3 415 432 412 585 600 588 795 813 743 739 813 798 637 658 618 535 541 568 382 410 409 348 332 349 167 170 170 528 535 536 Table C-6: Measured turbidity downstream, test 5. Elapsed Time (min) 5 10 15 20 25 30 35 40 NTU 1 413 353 455 277 616 751 461 460 TRIAL 1 NTU 2 397 376 462 302 562 642 487 431 NTU 3 391 355 424 293 586 693 495 438 TRIAL 2 NTU 1 NTU 2 NTU 3 458 463 468 417 424 388 528 537 502 443 448 473 628 586 595 724 735 721 464 462 478 566 552 589 158 45 556 571 521 Table C-7: Measured turbidity downstream, test 6. Elapsed Time (min) 5 10 15 17.5 20 25 30 35 40 45 50 55 60 65 70 75 80 NTU 1 702.1 670.2 684.9 565 528.3 471.2 523.7 479.2 844.9 482.8 286.8 272.9 239.8 269.3 287.3 346.2 450 TRIAL 1 NTU 2 717.4 671 700.9 553.3 517.2 461.4 512.7 478.7 831.2 492.2 295.3 270 237.9 260.2 286 348.2 447.3 NTU 3 656.8 682.2 693.1 546.6 540.6 474.9 523 474.3 826.9 478.5 291.2 280.1 240.6 265.2 288.8 337.3 447.6 TRIAL 2 NTU 1 NTU 2 NTU 3 659.4 619.5 485.8 480.3 569.3 527.3 660.3 556.8 617.9 456.2 297.6 210.4 334.3 251.6 358.2 397.1 378.2 661.7 624.7 477.7 508.3 580.7 509.1 641.3 569.2 566.5 478.4 296.1 207.9 326.8 251.7 356.5 402.8 374.4 652.7 612.5 480.9 537.2 594.8 552.5 637.7 566.3 601.3 488.7 291.4 204.7 332 254.4 360.7 406.3 382.2 159 APPENDIX D: NUTRIENT RESULTS Nutrient measurements are shown in the following tables. A table is included for each of the four nutrients tested. Measurements are sorted by test number and sample. Test numbers 1, 5 and 9 correspond to nutrients measured after the first, fifth and ninth liter of DI rinse water was passed through the samples, respectively. Test numbers 10 and 11 correspond to filtered test water and one rinse of DI water after the test water, respectively. Test N is the prepared test water. Samples A through E are 18" sock compost, 12" sock compost, cypress mulch, hardwood mulch and straw, respectively. Check tests are prepared solutions used to verify the performance and consistency of the nutrient test methods. Table D-1: Measured Nitrite (NO2) by sample. Test # Sample 1 A 1 B 1 C 1 D 1 E check1 0.32 5 A 5 B 5 C 5 D 5 E check2 0.2 check6 0.32 9 A 9 B 9 C 9 D 9 E Trial 1 Measured NO2 (ppm) Trial 2 Measured NO2 (ppm) Test # Sample 0.400 0.420 0.040 0.520 -0.080 0.311 4.700 0.740 0.035 0.215 0.078 0.184 0.310 4.900 0.235 0.020 0.080 0.068 0.400 10 A 0.500 10 B 0.170 10 C 0.480 10 D 0.040 10 E 11 A 11 B 0.700 11 C 0.040 11 D 0.225 11 E 0.080 check5 0.4 N A N B N C 0.220 N D 0.025 N E 0.078 0.068 Trial 1 Measured NO2 (ppm) 0.980 0.340 0.172 0.002 0.044 1.890 0.120 0.036 0.038 0.068 0.378 0.474 0.490 0.480 0.490 0.500 Trial 2 Measured NO2 (ppm) 2.300 0.315 0.040 0.065 0.105 1.900 0.145 0.035 0.065 0.075 160 check3 0.4 0.393 Table D-2: Measured Ortho-phosphate (P) by sample. Test # Sample Trial 1 Measured P (ppm) Trial 2 Measured P (ppm) Test # Sample Trial 1 Measured P (ppm) Trial 2 Measured P (ppm) 1 1 1 1 1 check 5 5 5 5 5 check6 check 9 9 9 9 9 check2 A B C Db Eb 0.1ppm A B C D E 0.08ppm 0.05ppm B B C D E 0.2ppm 2.600 0.600 1.050 26.500 34.000 0.094 9.500 0.890 0.068 0.478 3.025 0.055 0.049 9.800 0.275 0.043 0.153 2.130 0.173 2.400 0.500 0.940 24.520 31.720 0.925 0.070 0.480 3.060 0.285 0.045 0.173 2.135 10 10 10 10 10 11 11 11 11 11 check5 N N N N N A B C D E A B C D E 0.2ppm A B C D E 2.280 0.276 0.314 0.284 2.856 4.790 0.188 0.136 0.208 1.740 0.174 0.510 0.525 0.530 0.525 0.545 5.300 0.240 0.320 0.335 2.775 24.500 0.215 0.145 0.260 1.745 161 Table D-3: Measured Nitrate (NO3) by sample. Test # 1 1 1 1 1 check check 5 5 5 5 5 check check 9 9 9 9 9 check check Sample Trial 1 Measured NO3 (ppm) Trial 2 Measured NO3 (ppm) A B C D E 0.4ppm 1ppm A B C D E 0.4ppm 0.6ppm A B C D E 1ppm 0.4ppm 2.2400 2.6600 -0.0280 -0.0560 -0.1320 0.4550 -0.9033 0.4800 0.0200 -0.0300 -0.0340 0.3500 -1.7575 0.1400 0.0450 -0.0250 -0.0775 1.0610 -- 1.6000 1.1000 -0.5600 0.0800 -0.0800 -1.1040 1.0300 0.7150 -0.0500 -0.0600 -0.0317 -0.6360 1.0700 0.2050 0.0125 -0.0325 -0.0438 -0.3330 Test # 10 10 10 10 10 check 11 11 11 11 11 N N N N N Sample Trial 1 Measured NO3 (ppm) Trial 2 Measured NO3 (ppm) A B C D E 1ppm A B C D E A B C D E 0.4600 0.2240 0.2580 0.0840 0.0000 0.8660 0.8240 0.1000 0.0300 -0.0020 0.1040 0.5060 0.5270 0.4740 0.5260 0.4800 0.3600 0.1960 0.0767 0.0420 0.0180 -1.4800 0.1760 0.0440 0.1120 -0.0280 ------ 162 Table D-4: Measured Ammonia (NH3) by sample. Test # Sample NH3-N Ion Rdg #1 NH3-N Ion Rdg #2 Test # Sample NH3-N Ion Rdg #1 NH3-N Ion Rdg #2 1 A 1.7500 1.3800 10 A 3.5600 7.3600 1 B 0.3300 0.4980 10 B 0.4220 0.4000 1 C 0.1400 0.1810 10 C -- -- 1 D 0.0048 0.0028 10 D 0.1530 0.2180 1 E 0.0150 0.1230 10 E 0.5400 0.7400 5 A 1.2600 1.3400 11 A 1.7600 0.9510 5 B 0.1900 0.3600 11 B 0.1900 0.1810 5 C 0.4810 0.6900 11 C 0.0431 0.5000 5 D 0.2870 0.1050 11 D 0.0352 0.2200 5 E 0.6840 0.1250 11 E 0.0724 0.2870 9 A 2.3600 2.9100 check1 0.3 0.414 -- 9 B 0.7480 1.5600 check2 0.5 0.451 -- 9 C 0.1750 0.4440 check3 0.7 0.684 -- 9 D 0.2310 0.3660 check4 1 0.74 -- 9 E 0.1350 1.1300 check5 2 1.24 -- check6 2.5 1.88 -- check7 3 2.35 -- check8 5 4 -- 163 APPENDIX E: METAL RESULTS Measured metal concentrations for the prepared metal solution and filtered samples are shown according to test number and filter material. Aliquots prepared for testing included a known concentration of Yttrium. The percent Yttrium recovered during testing is shown for each sample measured. Measured results are corrected for volume reductions (aliquot size). Measurements are an average of three test iterations. Table E-1: Measure metal concentrations. Material 18" Compost sample 12" Compost sample Cypress mulch Hardwood mulch Straw Material 18" Compost sample 12" Compost sample Cypress mulch Hardwood mulch Straw Material 18" Compost sample 12" Compost sample Cypress mulch Hardwood mulch Straw M Prepared metal solution (ppm) Cu 0.479 0.467 0.468 0.459 0.481 Zn 0.455 0.464 0.467 0.452 0.466 Pb 0.456 0.463 0.464 0.459 0.468 Test 12 Filtered metal solution (ppm) Cu 0.002 0 0.053 0.002 0.236 Zn 0.016 0.012 0.095 0.018 0.314 Pb 0.002 0 0.035 0.012 0.218 Test 13 1L DI rinse after metal test (ppm) Cu 0.014 -0.008 -0.008 -0.01 0.01 Zn 0.028 0.004 0.004 0.006 0.022 Pb 0.002 -0.002 0.002 0.002 0.012 Yttrium recovery (%) 103.6 101.4 101.8 99.2 96.9 Yttrium recovery (%) 104.3 105.3 103.5 100.8 104.0 Yttrium recovery (%) 103.9 100.3 102.8 101.3 104.8 164 REFERENCES Aitani, Abdullah M. (2006). Propylene Production. Encyclopedia of Chemical Processing. 2461-2166. ASTM D4491, 2009, "Standard Test Methods for Water Permeability of Geotextiles by Permittivity," ASTM International, West Conshohocken, PA. 2009. ASTM D4751, 2012, "Standard Test Method for Determining Apparent Opening Size of a Geotextile," ASTM International, West Conshohocken, PA. 2012. ASTM D5141, 2009,"Standard Test Method for Determining Filtering Efficiency and Flow Rate of the filtration component of a Sediment Retention Device Using SiteSpecific Soil," ASTM International, West Conshohocken, PA. 2009. ASTM D7351, 2007, "Standard Test Method for Determination of Sediment Retention Device Effectiveness in Sheet Flow Applications," ASTM International, West Conshohocken, PA. 2007. Bare, Jane C., Gregory A. Norris, David W. Pennington & Thomas McKone. (2003). TRACI: The Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts. Journal of Industrial Ecology. 6(3-4), 49-78. Barrett, Michael E., Joseph F. Malina, Jr., & Randall J. Charbeneau. (1998). An evaluation of geotextiles for temporary sediment control. Water Environment Research. 70: 283-290. Beighley, R. Edward & Julio R. Valdes. (2009). Slope Interrupter Best Management Practice Experiments on a Tilting Soil Bed with Simulated Rainfall. Journal of Irrigation and Drainage Engineering. 135(4): 480-486. Benjamin, Mark M. (2002). Water Chemistry. New York, NY: The McGraw Hill Company, Inc. Berrittella M., A. Certa, M. Enea & P. Zito. (2007). An Analytic Hierarchy Process for the Evaluation of transport Policies to Reduce Climate Change Impacts. Fondazione Eni Enrico http://www.feem.it/Feem/Pub/Publications/WPapers/default.htmBhatia, S., & Huang, Q. (1995). Geotextile filters for internally stable/unstable soils. Geosynthetics International, 2(3), 537565. California Department of Transportation (CALTRANS). (2002) Caltrans Construction Site Runoff Characterization Study: Monitoring Seasons 1998 2002. Report CTSW-RT-02-055. 165 Carter, D.L, C.E. Brockway, K.K. Tanji. (1993) Controlling Erosion and Sediment Loass from Furrow-Irrigated Cropland. Journal of Irrigation and Drainage Engineering. 119(6), 975-988. Chang, J. S., & Vigneswaran, S. (1990). Technical note ionic strength in deep bed filtration. Environmental Engineering, 24(I), 3-8. Chang, Ni-Bin, Mary Wanielista, Fahim Hossain & Lisa Naujock. (2008) Material Characterization and Reaction Kinetics of Green Sorption Media for Nutrients Removal. Environmental and Water Resources Congress 2008: Ahupha's. Curran, Mary Ann. (1996). Environmental Life-Cycle Assessment. New York, NY: McGraw-Hill. Daniel, T.C., P.E. McGuire, D. Stoffel, & B. Miller. (1979). Sediment and nutrient yields from residential construction sites. Journal of Environmental Quality. 8(3): 304308. Demars, Kenneth R., Richard P. Long, & Jonathan R. Ives. (2004). Erosion Control Using Wood Waste Materials. Compost Science & Utilization. 12(1): 35-47. Downs, H.W. & R.W. Hansen. (1998). Estimating Farm Fuel Requirements. Retrieved from http://www.ext.colostate.edu/pubs/farmmgt/05006.html on April 22, 2013. Droste, Ronald L. (2005). Theory and Practice of Water and Wastewater Treatment. New York, NY: John Wiley & Sons, Inc. Faucette, L.B., J. Governo, R. Tyler, G. Cigley, C.F. Jordan, & B.G. Lockaby. (2009a). Performance of compost filter socks and conventional sediment control barriers used for perimeter control on construction sites. Journal of Soil and Water Conservation. 64(1): 81-88. Faucette, L.B., F.A Cardoso-Gendreau, E. Codling, A.M. Sadeghi, Y.A. Pachepsky, & D.R Shelton. (2009b). Storm Water Pollutant Removal Performance of Compost Filter Socks. Journal of Environmental Quality. 38: 1233-1239. Faucette, L.B., B. Scholl, R.E. Beighley, & J. Governo. (2009c). Large-Scale Performance and Design for Construction Activity Erosion Control Best Management Practices. Journal of Environmental Quality. 38: 1248-1254. Finnveden, Gran. (1999). A Critical Review of Operational Valuation/Weighting Methods for Life Cycle Assessment. Swedish Environmental Protection Agency, Report 253. Stockholm, Sweden. Fruehan, R.J., O. Fortini, H.W. Paxton, R. Brindle. (2000). Theoretical Minimum Energies to Produce Steel for Selected Conditions. Retrieved from http://www1.eere.energy.gov/manufacturing/resources/steel/pdfs/theoretical_mini mum_energies.pdf. 166 Georgia Environmental Protection Division (GaEPD). (2010). Water Quality in Georgia 2008-2009. Atlanta, GA: GaEPD, Georgia Department of Natural Resources. GaBi (version 6.0) [software]. PE International. Georgia Soil and Water Conservation Commission (GSWCC). (2000). Manual for Erosion and Sediment Control in Georgia. Athens, GA. Giroud, J. P. (2006). Geosynthetics engineering: successes, failures and lessons learned. Terzaghi lecture notes. (September 18, 2006). Geosyntec Consultants & Wright Water Engineers, Inc. (2009). Urban Stormwater BMP Performance Monitoring. Gregory, John. (1975). Interfacial Phenomena. The scientific basis of filtration: proceedings of the NATO Advanced Study Institute held at Cambridge, U.K., July 220, 1973. Leyden: Noordhoff. 1975. Guine, Jeroen R. (2002). Handbook on Life Cycle Assessment: Operational Guide to the ISO Standards. Dordrecht, The Netherlands: Kluwer Academic Publishers. Harbor, Jon. (1999). Engineering geomorphology at the cutting edge of land disturbance: erosion and sediment control on construction sites. Geomorphology, 31: 247-263. Ho, Carlton L., Kevin C. Wong, & Justin P. Seltzer. (2011). Engineering Characteristics of Compost used for Erosion and Sedimentation Control. Geo-Frontiers 2011 (pp. 12551265). Hoare, David J. (1982). Synthetic fabrics at soil filters: A review. Journal of Geotechnical and Geoenvironmental Engineering, 108(GT10), 1230. Honjo, Yusuke, & Danieie Veneziano. (1989). Improved Filter Criterion for Cohesionless Soils. Journal of Geotechnical Engineering, 115(1), 7594. Intergovernmental Panel on Climate Change (IPCC). (2007). Climate Change 2007: Working Group I: The Physical Science Basis. Retrieved from http://www.ipcc.ch/publications_and_data/ar4/wg1/en/ch2s2-10-2.html. Ison, C. R., & J.K. Ives. (1969). Removal mechanisms in deep bed filtration. Chemical Engineering Science, 24(4), 717-729. Ives, K. James. (1975) Capture mechanisms in filtration. The scientific basis of filtration: proceedings of the NATO Advanced Study Institute held at Cambridge, U.K., July 220, 1973. Leyden: Noordhoff. 1975. 167 Johansson, Jessica. (1999). A Monetary Valuation Weighting Method for Life Cycle Assessment Based on Environmental Taxes and Fees. Master Thesis, Department of Systems Ecology, Stockholm University, Sweden. Keener, Harold M., Britt Faucette, & Michael H. Klingman. (2007). Flow-through Rates and Evaluation of Solids Separation of Compost Filter socks versus Silt Fence in Sediment Control Applications. Journal of Environmental Quality. 36: 742-752. Kim, Hunho, Eric A. Seagren & Allen P. Davis. (2003). Engineered Bioretention for Removal of Nitrate from Stormwater Runoff. Water Environment Research. 75(2003): 355 367. McCaleb, M.M. & R.A. McLaughlin. (2008). Sediment trapping by five different sediment detention devices on construction sites. Transactions of the ASABE. 51(5): 1613-1621. McFalls, Jett, Ming-Han Li, & Aditya B. Raut Desai. (2009). Roadside Sediment Control Device Evaluation Program: Technical Report (Report Number FHWA/TX-10/05948-1). Austin, TX: Texas Department of Transportation, Research and Technology Implementation Office. Retrieved from http://tti.tamu.edu/documents/0-5948-1.pdf Montero, C. M. & L.K. Overmann. (1990). Geotextile Filtration Performance Test. Geosynthetic Testinq for Waste Containment Applications, ASTM STP 1081. American Society for Testing and Materials, Philadelphia, 1990. National Pollutant Discharge Elimination System (NPDES). (2012). General Permit for Discharges from Construction Activities. Palamutcu, S. (2010). Electric energy consumption in the cotton textile processing stages. Energy. 35(2010), 2945-2952. PE INTERNATIONAL. (2012). Database Upgrade 2012: Updates and improvements. PR Consultants. (2010). Introduction to LCA with SimaPro 7. Rickson, R. J. (2006). Controlling sediment at source: an evaluation of erosion control geotextiles. Earth Surface Processes and Landforms, 31: 550560. Santamarina, J. Carlos. & A.K. Klein. (2001). Soils and waves: Particulate Materials Behavior, Characterization and Process Monitoring. Chichester, England: John Wiley & Sons, Inc. Sharma, Hari D. & Krisha R. Reddy. (2004). Geoenvironmental Engineering. Hoboken, NJ: John Wiley & Sons, Inc. 168 Scientific Applications International Corporation (SAIC). (2006). Life Cycle Assessment: Principles and Practice (Report Number EPA/600/R-06/060). Cincinnati, OH: USEPA, Office of Research and Development. Site, Alessandro Delle. (2001). Factors Affecting Sorption of Organic Compounds in Natural Sorbent/Water Systems and Sorption coefficients for Selected Pollutants. A Review. Journal of Physical and Chemical Reference Data. 30(1)187 439. Stevens, Ellen, Billy J. Barfield, S.L. Britton & J.S Hayes. (2004). Filter Fence Design Aid for Sediment Control at Construction Sites (Report Number EPA/600/R04/185). Cincinnati, OH: USEPA, Office of Research and Development. Storey, Beverly B., Aditya B. Raut Desai, Ming-Han Li, Harlow C. Landphair & Timothy Kramer. (2005) Water Quality Characteristics and Performance of Compost Filter Berms (Report Number FHWA/TX-06/0-4572-1). Austin, TX: Texas Department of Transportation, Research and Technology Implementation Office. Retrieved from http://tti.tamu.edu/documents/0-4572-1.pdf. Terzaghi, K. & R. B. Peck. (1967). Soil mechanics in engineering practice. 2nd edition. New York: Wiley. Texas Department of Transportation (TXDOT). (2004). Storm Water Field Inspector's Guide. Austin, TX: TXDOT, Environmental Affairs Division. Theisen, M.S. (1992). The role of Geosynthetics in Erosion and Sediment Control: An Overview. Geotextiles and Geomembranes. 11(1992): 535-550. United States Environmental Protection Agency (USEPA). (1971). "Test Method 160.2:Residue, Non-Filterable (Gravimetric, Dried at 103-105C)," Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (1971). "Test Method 354.1: Nitrogen, Nitrite (Spectrophotometric)," Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (1971). "Test Method 352.1: Nitrogen, Nitrate (Colorimetric, Brucine)," Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (1971). "Test Method 365.2: Phosphorous, All Forms (Colorimetric, Ascorbic Acid, Single Reagent)," Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (1974). "Test Method 350.3: Nitrogen, Ammonia (Potentiometric, Ion Selective Electrode)," Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (1974). "Test Method 180.1: Turbidity (Nephelometric)," Washington, DC: USEPA. 169 United States Environmental Protection Agency (USEPA). (1974). "Test Method 200.7: Determination of Metals and Trace Elements in Water and Wastes by Inductively Coupled Plasma Atomic Emission Spectrometry," USEPA Report 600/R-94111. Washington, DC: USEPA. United States Environmental Protection Agency (USEPA). (2002). Development Document for Proposed Effluent Guidelines and Standards for the Construction and Development Category. Washington, DC: USEPA, Office of Water. United States Environmental Protection Agency (USEPA). (2010). Straw or Hay Bales. Retrieved March 25, 2013, from http://cfpub.epa.gov/npdes/stormwater/menuofbmps/index.cfm United States Environmental Protection Agency (USEPA). (2012a). Stormwater Best Management Practices: Silt Fences. Washington, DC: USEPA, Office of Water. Retrieved online at: http://www.epa.gov/npdes/pubs/siltfences.pdf. United States Environmental Protection Agency (USEPA). (2012b). Fiber Rolls. Retrieved March 25, 2013, from http://cfpub.epa.gov/npdes/stormwater/menuofbmps/index.cfm United States Environmental Protection Agency (USEPA). (2012c). Tool for the Reduction and Assessment of Chemical and Other Environmental Impacts (TRACI). Retrieved from http://www.epa.gov/nrmrl/std/traci/traci.html. The Weather Channel. (2012). Monthly Averages for Macon, GA. Retrieved from http://www.weather.com/outlook/recreation/outdoors/wxclimatology/monthly/gra ph/USGA0346?from=search Welsch, D.J. (1991). Riparian Forest Buffers. NA-PR-07-91, U.S. Department of Agriculture, Forest Service, Northeastern Area. Radnor, PA. Xiao, Ming & Lakshmi N. Reddi. (2000). Comparison of Fine Particle Clogging in Soil and Geotextile Filters. Advances in Transportation and Geoenvironmental Systems Using Geosynthetics (pp. 176185). American Society of Civil Engineers. Zech, W. C., J.L Halverson & T.P Clement. (2008). Intermediate-Scale Experiments to Evaluate Silt Fence Designs to Control Sediment Discharge from Highway Construction Sites. Journal of Hydrologic Engineering. 13(6): 497-504. Figure 85 170